Your Lying Eyes: What Klinsmann Sees In Orozco-Fiscal

TSG’s Tuesday plays contrarian–or does he–on the maligned US centerback

Orozco-Fiscal, popping up at every Klinsmann camp. (Credit: Matt Mathai)

At TSG, we’ve tried to offer some explanation for Michael Orozco-Fiscal.

All the pundits, professional and armchair men alike seem to be in agreement. They sing a holiday chorus asking for this weak brew to be left out in the cold in favor of the stout, USMNT-eligible MLS defenders of the moment. Yet, Jurgen Klinsmann continues to select the Primerican.

What’s he seeing that the rest of us don’t?

We’ve looked at his individual statistics and they don’t seem impressive.

But soccer and statistics share a tenuous relationship without the help of Hadoop. What happens on the soccer field isn’t always easily quantified. A shot is a shot, on target or not; a corner is a corner; an offside is an offside, most of the time; a save is a save; and a goal is usually a goal. On the basics, our eyes can agree.

At the same time we’re citing a player’s passing statistics to illuminate their performance, youth soccer coaches around the country are reminding the next generation that it takes two to make a pass.

Xavi would hardly be sporting the same Tom Brady-level passer rating if it weren’t for his teammates moving dizzily around him like electrons in orbit around a proton, buzzing into pockets of space. The Messi-Welker connection sadly does not work here.

The basic measures don’t do much when we’re trying to assess a defender.

Good defending comes from a team working like a finely tuned machine, not just from one-on-one tackles. Take this statement by Jay DeMerit to TSG last year:

On ball defending is about two things. Desire and effort in that the attacker coming at you is not getting by and second about the responsibility to your teammates not to let that player get by you and put them under pressure because you did your job.

Individual numbers lie in soccer because soccer is all about how a team of players function together.

Standout individual performances can sometimes mean the team is broken. Sometimes it makes more sense to assess the team’s performance with and without a certain player to understand the impact he makes.

So we come to the US Men under Jurgen Klinsmann and his quest to transform American soccer.

We can all argue about what we see happening on the field until we’re blue in the face. Can some team numbers tell us something more about this conundrum?

The argument for Orozco-Fiscal is that he (or a player of his type) is crucial to the success of Klinsmann’s project by allowing the US to play a higher defensive line, win the ball back more quickly and keep possession more effectively by reducing the space between their lines.

The argument against is that he’s a defensive liability whose mistakes give the opponent too many chances to score.

Under Jurgen Klinsmann, the USA has averaged the following per 45 minutes:

———-

US opponents have averaged the following per 45 minutes:

Source: MLSSoccer.com Chalkboards
* Passing statistics exclude USA vs Mexico on 8/10/2011 as data was not available.

If we compare how the USA team performs by these measures with Orozco-Fiscal compared to without him on the field:

  • The US averages 70.1 more attempted passes, 68.3 more successful passes and a completion rate 4.5% higher per 45 minutes with Orozco-Fiscal on the field.
  • The US averages 0.5 fewer shots and 0.7 fewer shots on target with 0.4 more goals scored.
  • US Opponents average 39.4 fewer attempted passes, 35.4 fewer successful passes and a completion percentage 3.2% lower with Orozco-Fiscal on the field.
  • They average 0.4 fewer shots but 0.2 more shots on target with no difference in goals scored. The conversion rate of shots to goals was 0.9% higher.

What do these numbers–with a Bayesian lens–mean?

Team d and possession?

Klinsmann’s team keeps possession far better with Orozco-Fiscal on the field and the score more goals despite creating a few less chances.

The USA wins the ball back more quickly so their opponents have less of it and create fewer shooting opportunities. Whenever a team plays a high line, the chances they concede are of better quality, so we see the opponent’s shots are somewhat more likely to be on target, and a slightly higher conversion rate of shots to goals.

What the US gains by keeping the ball away from their opponents attack mitigate the impact of his individual defensive shortcomings so they don’t lessen the quality of the team’s overall defending, at least over the course of these 4 matches.

When seen within the context of team’s performance, Orozco-Fiscal isn’t quite  the defensive liability that he is sometimes singled out as based on his individual stats. And 70 more American passes per half when he plays is good evidence for his defense. (Or is it offense?) His presence allows the US to play a possession style that is difficult to maintain when a deeper defensive line dictates more vertical space between the lines. The key, as always, remains turning that additional possession into quality scoring chances.

At the very least, this seems to show what Klinsmann sees in a player that many fans and pundits have declared to be simply “not good enough.” His apparent impact on the team’s ability to keep possession, Klinsmann’s oft-stated goal for his side, is the reason he has continued to be named to the roster for international friendlies, much to the pundits chagrin. Maybe they’re just not seeing the big picture.

Is Orozco-Fiscal’s value to the whole greater than a single poor defensive play–or in this sense–a single poor defensive chance observation? Well, that’s subjective. And that is of course, the manager’s decision.

For now, at least, Orozco-Fiscal is the future.

About these ads

98 responses to this post.

  1. Posted by Michael on 2011/11/08 at 10:26 AM

    Fascinating analysis.

    Question:
    Is there direct evidence that the US keeps a higher line when Orozco-Fiscal is on the field? Are there heat-maps available to back that idea up?

    Reply

    • I’m not finding any through google, but they may be buried somewhere I haven’t looked.

      Reply

      • Posted by Sgc on 2011/11/08 at 12:02 PM

        It would be manually intensive (you’d have to generate the ‘heat map’ part yourself), but you can highlight only parts of the game under the chalkboard feature, and do a combined selection of both CBs of everything they did. That would basically give you the picture, I would think.

        If you basically superimposed all the dots (though it would be helpful to get the vector lines out of there) of both CBs when Orozco is in versus both when he’s out, you might possibly see something.

        Reply

    • If you look at the chalkboards, yes. I’ve mentioned Honduras’s passing stats in the final third in first half vs second half as more evidence in a comment below.

      Reply

  2. Posted by Kevin O' on 2011/11/08 at 10:35 AM

    I will hold my Klinsmann criticisms until after January, here’s hoping we see a plethora of new defenders.

    I’m more excited about the U23′s at the moment.

    Reply

  3. Posted by obxfly on 2011/11/08 at 10:47 AM

    It would be interesting to see what other elements of the team changed at or near the same time he was not present on the field. For example, was another player pulled off about the same time as him? This would also affect these stats.

    Reply

    • Yeah, it seems a bit misleading to single out Fiscal as the source of the better possession when he’s also playing with the other starters. Although, some of those subs (Bradley) are theoretically supposed to lend themselves to more possession.

      Reply

  4. Posted by Matt Mathai on 2011/11/08 at 10:47 AM

    Excellent. Gives me something to think about.

    Reply

  5. Posted by Rob on 2011/11/08 at 10:49 AM

    Does it have anything to do with him typically starting , and thus, playing with the starters rather than all of the substitutes?

    Reply

  6. For statistical analysis to be valuable you must have a large sample size. Under Klinsmann the US has played, what, 5 games? How can you produce meaningful statistics from such a small sample? What about other variables like the rest of the lineup and the opponent…?

    Reply

    • Posted by matthewsf on 2011/11/08 at 11:02 AM

      Actually–that’s incorrect. Smaller than “statistically significant” is “directional” with a probability of accuracy–that’s how we hoped these stats came across. You are right — it is however — a disclaimer on the piece.

      Reply

    • This isn’t a robust statistical analysis – it’s just an attempt to get some insight using what few numbers we do have. We could control for opponent quality, playing style, set variables for players on the pitch to see what the significance of each player but, as you suggest, there’s not really much point in going through that kind of analysis with such a small sample size.

      Nonetheless, the numbers suggest an interesting pattern that’s worth talking about.

      Reply

    • Also if you’re gonna cite sample size.. we’ve only seen Fiscal play in five games for the US, how can we possibly have any idea what kind of player he is in such a small sample size? We can’t definitively. There’s no way. Anything we would do is a best guess, and this is an objective (stats-based) best guess compared to a subjective (what I can see) one at least. Very interesting, although I’m not sure I’m surprised.

      Reply

  7. A good argument, well-researched and backed with facts, for keeping Orozco-Fiscal in the fold for the moment.

    I think that the gut reaction most USMNT fans would have to having him on the field is that he’s a liability. Your numbers don’t totally hold him up to be a messiah at the back, but they show that undoubtedly they fit in much better with Klinsy’s game than other options.

    The USMNT is a work in progress. Good thing we haven’t even started WCQ yet.

    Reply

    • Posted by Jared on 2011/11/08 at 11:57 AM

      Do we know that he fits in better than other options? I think a guy like Cameron at Houston would be an even better fit for that role because he actually is a good MLS defender unlike Orozco. It feels to me like Klinsmann is just sort of going along with the old guard except for Orozco without trying any other options. I’d love to know how he decided on Orozco as the speed guy.

      Reply

  8. Posted by Jared on 2011/11/08 at 11:07 AM

    Plaudits for a thought-challenging analysis, but I can’t help but think the key ingredient to any statistical study is missing here: sample size. I can’t imagine five matches producing near enough data to draw any real conclusions – particularly given the fact these have all been friendlies with rotating casts.

    That said, this is certainly data to watch moving forward, assuming – which I think we all are – Ozoco Fiscal continues to get a lion’s share of the RCB pie.

    Reply

    • Yet we’re constantly drawing subjective conclusions about a defender from a single mistake on one defensive play. Ask Tim Ream.

      Reply

      • Posted by Jared on 2011/11/08 at 11:52 AM

        That’s a fair point, though I’m not sure four matches, with all the confounding variables present, gives us a ton more than subjective observations of positioning, distribution, etc.

        For me, it’s more just frustration with the fact others aren’t being given a fair shake. Sure, the small amount of data we have available says MOF has been better than he’s being given credit for and better than his replacements. That doesn’t mean it says we should just pen him in as the starting RCB, I don’t think – especially given that we’re just looking at Orozco Fiscal vs. replacements rather than Orozco Fiscal vs. Onyewu and Orozco Fiscal vs. Goodson. Feels a little cherry picky to me, no?

        I’m a big fan, follower and consumer of the statistical revolutions in baseball, basketball and football, so I’m certainly sympathetic to your goal to dig in beyond what our eyes and a few gaffes tell us. I’m personally just of the opinion there’s not enough here to draw strong conclusions. I really like where it’s going, though.

        Semi-related: in baseball, defensive errors is a rather silly statistic because A) what is more important for a defender is his range, i.e. the number of balls he can get to compared to the rest of the league and B) errors tend to be over-dramatized, as there are so many plays in baseball that one error, in the grand scheme of things, isn’t generally all that destructive.

        In soccer though, where one singular game is generally more important than one singular baseball game, and where one goal conceded because of your error is generally much more destructive than one or two runs scoring because of your error (runs are much easier to score than goals), what is the break even point for a defender’s version of errors vs. auxiliary skills (like the ones you listed in your study)?

        Reply

      • Posted by Jared on 2011/11/08 at 11:59 AM

        I’ve drawn my conclusions on Tim Ream on more than one game. He’s made two mistakes that led to goals for the USMNT that I can recall and he’s been very poor for the Red Bulls. There’s a lot more data for Ream not being ready than on Orozco being the answer.

        Reply

    • Posted by Gary on 2011/11/08 at 11:53 AM

      Isn’t the point of the article “What Klinsman is seeing/thinking”. If it is, and I think it is, than this is the largest “sample size” available. Its all we or Klinsman have to go on, the games which he has coached since taking over the reigns.

      My fear is that some are looking for reasons to dismiss the ideas, rather than genuinely considering if this could be Klinsman’s thought process. Just my opinion.

      Reply

      • Posted by SamT on 2011/11/10 at 10:23 AM

        This is a great point, although I would also point out that Klinsmann has a much larger sample size (and of a slightly different nature) when you add the many practice sessions that he and the coaching staff have observed.

        Nice job, Tuesday. Give me messy stats and some thoughtful commentary any day — rather than waiting for a large sample size to accumulate. You might get a better answer in the long run, but as Keynes noted, “In the long run we are all dead.”

        Reply

  9. Gonna have to disagree here.

    For starters, not enough data points for this sample to be truly representative.

    Second, Orozco Fiscal is not the only variable in play, and correlation does not equal causation. Quality of opponents, the rest of the lineup, travel, etc. all impact how successful the team is in executing, not just whether or not MOF is in the lineup.

    Consider this, for example: the Yanks didn’t have the best of games against Belgium, a match in which MOF did not feature. Does that mean that his absence was a large factor in that middling performance? Or does the fact that Belgium is a much better opponent than an less-than-full-strength Costa Rica or Honduras, the nearly 6,000 mile flight preceding the match, the limited rest between game days, and lineup changes have a much more significant impact?

    Admittedly, I don’t have exact statistics on some of your metrics, but looking simply at goals, shots, and shots on goal with recollection of the general gist of possession, the US was statistically better against Costa Rica than Belgium, with MOF playing in the first of those two matches. But look at the changes to the Belgium match: Steve Cherundolo came in for Edgar Castillo, for one. Do those stats support anything along the lines of Castillo making a significant contribution that merits re-selection? (Castillo also featured against Mexico while not playing against Ecuador, Belgium, and Honduras; I have a feeling a similar data analysis might work out in Edgar’s favor, as well, albeit perhaps to a lesser degree.) Or are there too many lurking variables that make such a leap impossible given the tiny sample size?

    I’m inclined to say the latter. Orozco Fiscal’s minutes have come against a Mexico team not fielding several of its stars, as well as Costa Rica and Honduras, whose squads were also missing several regulars. The matches he didn’t feature in, meanwhile, came against Belgium and Ecuador, both of which also happened to be the second match after just a couple days of rest and not insignificant travel (particularly in the case of Belgium) that can impact performance (again, another variable that must be taken into consideration). MOF has yet to play in the back end of a two-match break, which skews the data for a variety of reasons.

    As much as this is a noble attempt to explain what many of us have left as inexplicable, it doesn’t have robust enough data behind it to make it stand up. Now, if we could get a multivariable regression with all significant variables taken into account, that would be a little bit of a different story (though it would still suffer from having so few samples).

    Reply

    • Absolutely. Correlation is not causation. But the correlation is certainly there, albeit in this small sample. But in international soccer we’re always going to have a small sample.

      But let’s do this – against Honduras, the US had 40 more passes in the first half with Orozco Fiscal than they did in the second half without him. With the US defense playing deeper, Honduras completed 15 of 23 passes in the final third during the second half vs 8 of 19 in the first half.

      For me, the numbers are more about the impact of the team being able to play a higher line than it is about any one player.

      Reply

      • That reasoning underscores the presence of another variable potentially at play: fatigue. The US pressed the issue early on, only to drop off later, with Honduras consequently seeing more of the ball. There’s also the question of where these passes took place and the validity of some passing stats in general.

        Fatigue may or may not have played a significant part, but the bottom line— and my main point— is that there are many variables involved that impact the data that aren’t fully taken into consideration.

        As I said, I wonder if the data would come out favorably for Edgar Castillo under this method of analysis. If so, do we make the same conclusions?

        Reply

        • Posted by Jared on 2011/11/08 at 12:01 PM

          If the data comes out favorably for Castillo then all statistical analysis of the game of soccer should be thrown out the window.

          Reply

          • Posted by matthewsf on 2011/11/08 at 12:13 PM

            Why?

            Reply

            • Posted by Jared on 2011/11/08 at 12:42 PM

              Because Castillo was beyond awful in those games. The data should back that up and if it doesn’t then I’m not sure it can be fixed. There is certainly an argument as to whether Orozco is good enough or not but I don’t think you’d find anyone other someone who is just a contrarian stating that Castillo wasn’t awful.

            • I disagree – even if one player is truly awful, they can often be carried by the rest of the team as long as the team performance is solid. I mean, we sometimes won games with Bornstein playing, didn’t we? I think that happened to some extent in the games Castillo played – Boca did a great job covering for him when needed and had a couple of his best recent performances.

              Regardless, having Castillo in the line-up doesn’t change the way the team plays in the same way that MOF does in my subjective viewing.

          • Well it could. look at it this way so far in Klinsmanns career if you go by that criteria, you have a 50% increase in goals when Robbie Rodgers is on the field. Is this information interesting? Yes. Do I think we should sacrifice a roster spot because these statics look to favor MOF? probably not.

            Reply

        • This is soccer. There are an infinite number of variables we can try and quantify. That’s what we love about it. It’s like life.

          And to answer your question: No, because Castillo is rubbish.

          But seriously, it’s the center backs that set the offside line and together determine how deep their team defends. How deep a team defends determines how close defenders are to midfielders and midfielders are to attackers when they win the ball back. A fullback just doesn’t have that same impact. And we have better ones with the same speed, better ability on the ball and defensively than Castillo.

          That’s what these numbers are showing – the impact of a high defensive line on keeping possession. We had 69 more passes in the second half against Belgium that we did in the first. Why? Because we took the risk of playing a higher line to get back in the game.

          Reply

          • And I would argue that that success came not as a result of something in the back line, but rather the addition of a second defensive midfielder. Beckerman came on for Rogers at the half, giving Maurice Edu- who had a hard time covering all the ground in front of the back four in the first half by himself- some much needed assistance on both sides of the ball. Beckerman helped shore things up defensively and also- most importantly in the case of the metrics you have been using- helped maintain possession by lessening the burden on Mo to distribute and winning more balls back.

            To sum, not so much a high back line, but a retooled midfield that was more opponent-appropriate was the difference in the second half v. Belgium. Perhaps the latter permits more of the former? I won’t say. But were it not for one huge Tim Howard stop v. Honduras, we very well could still be winless in the Klinsmann era, lamenting the attempt at playing a high line with someone whose defensive abilities aren’t the highest in Orozco Fiscal.

            Ultimately, as you alluded to, this all comes down to our subjective interpretations, as we’ll never get enough data for a robust analysis. And in that regard, we are simply seeing the games differently (midfield or high back line? MOF or other factors? etc), so I suppose we’ll just have to agree to disagree for now.

            Regardless, I sincerely applaud the effort. I know it’s never easy to go against the current so to speak. It’s also always refreshing to see someone try to look at things differently, particularly when popular sentiment is seemingly so lopsided, which I try to do as well in my writing.

            The big question is, though: how does Klinsmann look at MOF’s performance? Would really be curious to hear a candid analysis from JK and/or his staff, though I won’t be holding my breath.

            Enjoying the banter, always good fun.

            Reply

  10. Wow. One of my favorite TSG posts of all time here. Really interesting stuff.

    Obviously sample size and other variables are a problem here – but a problem that can’t be addressed yet, until more games are played. Still fascinating.

    Reply

  11. Posted by CrazyMike366 on 2011/11/08 at 11:37 AM

    One more thing that helps the US keep a higher line with him on the field is his speed. Under Klinsmann, the defensive mid spot (usu. Beckerman) appears to be charged with containing and forcing wide, not disrupting and tackling as under Bradley. So if the ‘big defender’ (usu. Gooch) doesn’t stick the tackle on the high line or a guy making a run gets past a US fullback, you need a ‘quick defender’ (usu. Fiscal) with loads of speed to get back and head off the chance.

    If I’m not mistaken, it was on this very site that I saw a similar analysis about a month ago, yeah?

    As far as I’m aware of the only US guys that are playing at a reasonably high level in that ‘quick defender’ mold are Fiscal, Pearce, Cameron and maybe Parkhurst. Alfredo Morales might fit that role, but I’ve never seen him play so I wouldn’t know.

    I want to see Gonzalez called in as much as anybody else, but I just don’t think he’d be high enough on the pecking order at the moment to make an impact. He’s too similar in what he would do for the team to Onyewu, and Onyewu is bigger, stronger, and more experienced in that role. Gooch, Goodson, Gonzalez and John (and possibly John-Anthony Brooks if he breaks out in the Olympics like some pundits are suggesting) should be fighting over just 2-3 ‘Big defender’ spots on the WC’14 roster. Perhaps by 2014 Gonzalez will have passed by a then 33 year old Gooch? I think he’ll get his chance, but he needs to cool his jets and keep improving in the meantime.

    Reply

    • Posted by Eric on 2011/11/08 at 12:19 PM

      I don’t think you can include Parkhurst in that group. Parkhurst has extremely good positional sense which sometimes makes it look like he’s faster than he is. He’s actually only average at best in the speed department. However, like I said, positional sense can negate much of that weakness for him.

      I agree about Pearce and Cameron though being the other two “speed” options at centerback. Boca, Gooch, Goodson and Ream are all too slow. I will say that I for one haven’t given up on Ream and think Klins’ eventual idea is to pair a possession oriented centerback (like Ream) with that speed back. John and Gonzalez are two interesting possibilities as well, John moreso because he seems to have slightly more speed and a little more composure on the ball than Omar right now.

      The problem seems to be that the US typically develops the Gooch type centerback. That is, youth coaches love having that huge, imposing player at centerback who can win anything in the air. Part of this is because of the reliance and emphasis on set pieces in the US game. If you’re dealing with free-kicks and corners all the time then a slow, but huge 6’3″ back is going to be more appealing than the faster 5’11″-6’0″ centerback, even if the latter also has better possession ability. I think Klinsmann is dealing with the a lack of options in the player pool right now for the system he wants to run. The players who are there are either too young (Brooks, Ream (Who is only 24, and we should remember most centerbacks don’t hit their peak until their late twenties), etc..) or they’re questionable in quality but have one of characteristics needed for Klinsmann’s system (Orozco-Fiscal).

      Reply

      • Posted by Thrasymachus on 2011/11/08 at 3:46 PM

        Ream needs to do some work in terms of becoming a more consistently competent defender. That said, the notion that he is slow is simply not true.

        Reply

        • Posted by Eric on 2011/11/08 at 5:50 PM

          Never said Ream was ready now. I just said I think he could eventually become the starting centerback Klinsmann wants. Also, he may not be slow, but he’s certainly not fast enough to fall under the context of that “speed” back Klinsmann wants.

          Reply

  12. Like someone else posted, we all have players we like and don’t like. I’ve not been a Tim Ream fan since 3-4 games ago. Is that a consensus here, doubt it. I don’t know if Orozco-Fiscal is the best defender but right now he’s playing and being evaluated. Plain and simple.

    The data is not all that telling but it does provide some talking points. I can’t think of 1 game where any of his supposed errors led to a score. I know he fouled early in his first 2 games but after that, I’m not sure he alone gave up any goals. In the end, I like Boca and Onyewu the most but if for our formation and strategy Orozco plays, I can live with it. I have more concerns for the midfield right now.

    Reply

  13. Posted by Deuce4Prez on 2011/11/08 at 12:03 PM

    I find interesting that the biggest flaw cited is sample size. Yet most of the critics use the same sample size to argue O-F shouldn’t be invited into camp. Either we need more data (games) to have an accurate evaluation or we don’t. It doesn’t just cut one way.

    Reply

    • Posted by Jared on 2011/11/08 at 12:16 PM

      I agree, but I think the sentiment is more “What has MOF done to secure a lion’s share of the minutes over others?” This study doesn’t really do enough to answer that question, IMO. This is a time to try different guys out, and there isn’t, I don’t think, sufficient cause to continue featuring MOF over them. Not to say there are, beyond any doubt, better options, but how do we know without trying?

      Reply

      • Posted by mathmatics on 2011/11/08 at 12:51 PM

        Trying a guy out necessarily means giving him the lion’s share of the minutes though. Otherwise, as the sample-size folks are pointing out, you don’t really learn anything.

        I much prefer giving guys a sustained run out unless it’s _obvious_ they can’t hack it (Castillo, Rogers).

        That’s the only reason I’m not up in arms about Beckerman’s status as a starter. He’ll either fend off his competition for the DM spot, or he’ll succumb to it. I trust quality will win out in the end.

        Reply

  14. Posted by euroman on 2011/11/08 at 12:08 PM

    What a waste of space this article was. The reason O-F is getting used has everything to do with JK’s Mexican assistant coach.

    Reply

    • Posted by matthewsf on 2011/11/08 at 12:23 PM

      Would love to hear your rationale on this. Here’s a question: Name another RCB with multiple reps at the position who has both speed and possession ability.

      That would be Heath Pearce and he was called in twice until he came up lame both times.

      Take a look at the U-23 camp. Only 2 players are “Onyewu-types”

      Reply

  15. Posted by Union on 2011/11/08 at 12:08 PM

    Is John A Brooks quick enough to take on the Fiscal role?

    Reply

  16. Posted by Freegle on 2011/11/08 at 12:18 PM

    A valiant effort to quantify the unquantifiable. There are simply too many variables to draw any valid conclusions on this subject.

    If I take a laser level and place it on the ground and it is flat for as long as the eye can see, that doesnt mean the earth is flat. And, If TSG were so inclined, I’m sure they could find some numbers to manipulate into data that says that MOF is the soul reason for Klinsmann’s poor win-loss record so far.

    Tuesday didn’t take a side on MOF here he simply gave a shot at illustrating what Kilnsy might be thinking… But here’s the bottom line: Objective and subjective info need to both be used and right now, MOF doesnt appear to pass the subjective evaluations of ametuer or professional analysis, regardless of what the numbers say.

    Reply

    • Posted by mathmatics on 2011/11/08 at 1:10 PM

      If nothing else though, the piece illustrates that MOF’s defensive contribution could be more than what you see when he’s near the ball.

      It’s easy to watch MOF get outmuscled in a challenge (is that something he’s accused of doing?) and wonder why Omar can’t get a shot in there.

      It’s a lot less intuitive to credit MOF for our team-wide ability to shorten the field, cutting down an opposing midfielder’s space and time with the the ball.

      The statistics presented here aren’t conclusive of course, but MOF’s critics would be well served to at least acknowledge this rationale for his inclusion.

      Reply

  17. can anyone quickly give a fifa ranking of opponents with and with out MOF, cause that may skew the numbers…

    Reply

    • Posted by Eric on 2011/11/08 at 12:31 PM

      I’d be curious to see those rankings also but I would just say that bringing them into the discussion opens up the debate about how accurate they are as to determining a team’s actual strength. Just my opinion though.

      Reply

      • Posted by mbw on 2011/11/08 at 5:10 PM

        The avg. Elo rating for the opponents Orozco Fiscal played against* is 50, for those he didn’t play against, 39. In my opinion, these figures understate the gap considerably — Costa Rica (against whom MOF played 90 minutes) is ranked 49, but brought a B+ team; Belgium (MOF played 0) is 45 but is clearly much better than its ranking reflects.

        *excluding Mexico, data for which don’t appear in the table.

        Reply

  18. Posted by KickinNames... on 2011/11/08 at 12:42 PM

    I think we can safely put the sample size discussion to rest, huh? Sounds like we’ve got a bunch of Bill Jamesians who haven’t been fed in a while.
    Interesting hypothesis that Tuesday puts forward and in my reading he didn’t put it forward as anything other than an interesting collection of data (small though it may be…) that seems to point to a few interesting reasons that MOF continues to be included in JK’s schemes.
    As many have put forward more eloquently than I, his intent seems to be change the style to a higher d line that compresses the midfiield and drives MF pressure and thereby possession.
    One reading of that seems to indicate that when MOF plays the possession and passing numbers skew in the right direction(s). Using the subjective viewpoint of JK’s strategy and tactics and the small sample of objective data tuesday reached a pretty sound conclusion IMO. MOF deserves a few more looks in the role.

    I heavily disagree BTW with many who call for a boca-gooch combo as “the best” for the mnt now or going forward. That combo would go completely opposite to the attacking style that JK is attempting to instill. IMO

    Reply

    • Posted by Jared on 2011/11/08 at 1:09 PM

      Get all that, and clever Bill James dig. But the fact remains: others deserve a shot, and he’s not even calling them in. I’m not talking about Gonzalez – where’s Geoff Cameron? Michael Parkhurst? I’m smitten that MOF is better than we thought, but frustration with Klinsmann, I think, is fair given that seemingly deserving players aren’t getting a call. Again, I’m not arguing for MOF one way or the other, other than to say I don’t think he’s been so stellar we should stop trying different guys out. And I don’t think the results of this study support that argument either.

      Reply

      • I think speculation on the rationale for why those guys are not included would be another intereting article. ;-)

        Reply

      • I think there’s arguments against each of those.

        1) I think Geoff Cameron will be called in in January. You probably don’t want to call him in for his first Klinsmann camp when the MLS Cup was on the line.
        Plus, remember, he’s only been playing — thank you Dom Kinnear — for a short while at CB

        2) Parkhurst for all his positional awareness still just isn’t very fast or physical. It’s an unfortunate limitation. Maybe he’s improved in those areas or maybe not. I haven’t seen enough of him.

        Also, it’s not that–perhaps or maybe it is–that Klinsmann is giving undo time perhaps to Orozco-Fiscal. It’s that he represents what he wants at that position. So if you want to play–right now–with a highline and test that out.
        And test out your CBs venturing forward, and test out Dolo pushing very high without compromising your attacking central midfield (as happened under Bradley) than for good or bad, he’s the guy.

        Because you can’t do it with Omar right now.

        Reply

        • Posted by Jared on 2011/11/08 at 1:56 PM

          Agree on nearly all accounts. Just the principle of the thing for me. My argument, as I’ve stated in multiple places, isn’t necessarily what Orozco is or isn’t or what the competition is or isn’t, but just that Orozco seems to be favored over others for reasons I can’t explain. Yes, he has the attributes (fast, right footed, plays center back) Klinsmann is looking for, but are there really no others who have similar traits? Also, as much as Klinsmann is looking for someone with speed to cover overlapping outside backs and an aggressive foil in the middle, isn’t he also looking for someone who can pass the ball well out of the back, too? For me, and maybe the numbers say otherwise, MOF struggles with composure and distribution.

          I don’t like how Internet discussions tend to come off as confrontational. I think we all agree more than we disagree here, but it comes off differently as we try to illustrate our points. Great discussion here by all.

          Also: sucks Geoff Cameron is 26, no?

          Reply

      • Posted by Martin on 2011/11/08 at 8:25 PM

        It’s very hard to say JK isn’t trying other guys out when we are only five games into the experiment.

        How do we know that JK hasn’t said to himself “I’ll give Fiscal half a dozen games at this and then, I’ll give Cameron the next six games.” or something like that.

        It’s very clear JK likes to gives guys a run of games in a row to prove a point(See Beckerman), plus he has said that he prefers to do things that way.

        That only makes sense since both Fiscal and Cameron are newbies to the defense.

        Shuttling in a new guy every game would not be fair to Howard and Boca, who seem to the designated “new CB candidate evaluators” since they have played every game.

        Perhaps you should be more patient.

        As for Parkhurst, based on what I’ve seem of him in the past he would be more of a candidate along with Ream to replace Boca. This is a player who makes his money on positioning and anticipation not so much sheer physical ability.

        Reply

  19. I thought it was a well thought out and informative article. Thanks. I don’t think one has to agree with the selection of the player to see the possible rationale for his inclusion.

    I think it bares watching in the future as it could predict future camp match ups, call ins, and player pool scouting.

    Its interesting. Thanks.

    Reply

  20. Posted by Gregorio on 2011/11/08 at 1:26 PM

    Wow, great article and great commentaries. I can feel the engergy pulsating through the screen. Mr. Tuesday Kudos for a fine article although I do think that maybe we need to do an intervention on you to get you outside more often. Despite its flaws it was a very compelling and comprenhensive peice. Now come the buts….1) Are you like Jk’s propaganda man? 2)Does Size Matter? (Sample size you degenerates!) 3)How can I dislike him for emotional reasons disguised as facts if you write this stuff?
    Ok I just want to applaud you for a fine article, very thought provoking, and a possiblity of JK’s reasoning. Yet all is based on conjecture. I think that the items talking about defensive positioning is apt, use an basball analogy of a short stop who doesn’t make diving stops because he’s better positioned (or slow if youre Jeter!). Factors such as these or such as what he does to motivate others, protect others are all important as Jay Demerits quotes assert. But Is MOF good for the team as it is now? playing possession? High Line? yes but at can we get someone better? WHat the heck do, I know but I loved the article and the emotions it produced. I think MOF is adequate but I like Robbie Rogers and think he needs more time so take what I write in some perspective.

    Reply

  21. Posted by WatertownMA on 2011/11/08 at 1:29 PM

    Remember that Ghana-USA game round of 16 … of course you do. Take note what happens when both center backs lack pace. Yes, it was OT and everyone was tired. Nonetheless, at least one CB with speed may reduce the likelihood of a “Gyan” blowing by the defense.

    Reply

    • Posted by Gary on 2011/11/08 at 1:32 PM

      WHile I am for a “speed” guy as one of the cb’s, as I remember it, wasn’t strength the cause of that goal. Didn’t Gyan discard Boca like a rag doll when he made the challenge on the “50/50″ ball? Maybe I’m wrong, its been a while and I’m getting older.

      Reply

      • Posted by matthewsf on 2011/11/08 at 1:37 PM

        It was a fantastic play. And considered in the context of this discussion:

        1) It was a fantastic play
        2) If the US defense isn’t sagging maybe, but only maybe, that over-the-top ball is a lot more difficult
        3) Asamoah Gyan is completely inconsistent; it was a fantastic play.

        Reply

        • Posted by Alex on 2011/11/08 at 1:47 PM

          It was like Gyan decided not to be bored and do something. And a “weird” bounce caused Boca to look around for a ball that was nowhere near his line of site (that replay was so embarrassing-looking. Boca is staring up into the sky and the ball is like, a foot in front of him)

          Reply

      • Posted by WatertownMA on 2011/11/08 at 4:29 PM

        As painful as it is, I encourage you to revisit the replay. Strength did have an impact to create opening for a shot, however, there was a lot of space that could have been made up w/ speed:

        - speed coming from DeMerit to close down and help,
        - speed from Boca in closing down on Gyan farther away from the goal,
        - speed in recovering after Gyan bounced Boca off the tackle to block a shot

        p.s. I digress for a moment, but I don’t understand why Boca didn’t hold Gyan’s jersey to ensure he didn’t break free.

        Reply

        • Posted by Martin on 2011/11/08 at 8:08 PM

          I just watched a replay of that goal.

          While it certainly would have helped if either Boca or Jay D were quicker and faster, it seemed like more a case of splitting the defenders.

          Boca and Jay were more or less caught even with each other, the ball took a good bounce for Gyan, allowing him to split them and he wasted no time with it.

          I suspect Boca didn’t grab Gyan’s jersey because it was a very bang, bang kind of play and Gyan was already accelerating past Boca before he realised what was happening. That goal was all about Boca and Jay D getting caught poorly positioned.

          Reply

  22. Posted by dude on 2011/11/08 at 2:29 PM

    All well and good, but we’re looking for defenders who are fast, keep possession, and win the ball back?

    Geoff Cameron. I also have a feeling, once he matures and gains experience, Ethan White will be knocking on this door. He’s a beast athletically, and won a starters role by several last ditch saves with speed.

    Reply

  23. Posted by berniebernier on 2011/11/08 at 6:00 PM

    I dont think the point of the article was that MOF is a great CB. The takeaway is the benefit of playing a highline.

    This article showed data to support what I had been seeing on the screen. When we play Gooch and Boca we play a deeper line. The deeper line is necessary for defense but results in our 10 field players being spread out over more space making it harder to keep possession via short passes (not sure if its a cause or an effect but when we play a deep line we tend to boot the ball up field to Altidore wo is on an island.

    Now the high line has its issues (mainly MOF) but those who want to play a deeper line wtih Gooch and Boca shouldn’t complain when we hoof it up field.

    As an aside I prefer the high line as we still have a while before WC or even the hex and hopefully have another CB come along because I am scared at how slow a gooch boca pairing will be in 3 years.

    Reply

    • Posted by Matthew on 2011/11/08 at 7:46 PM

      This site is awesome. There is an SAT math-test writer that is turning this into a word problem now. “If MOF played in 230 minutes and Tim Ream played ….” Great stuff.

      Reply

  24. Posted by DC Pete on 2011/11/08 at 8:11 PM

    That was top-notch, Tuesday, a completely different and technical perspective on defending and the art of creating a squad of players versus putting the most famous or “talented” 11 guys out there.

    Reply

  25. Posted by fellaini's_fro on 2011/11/08 at 9:53 PM

    I applaud the article, the author, contributors and discussion on here for a player with the best new acronym for a US players since MB90. MOF.

    Trying to to guess at Klinnsy’s thinking regarding MOF using stats was a stroke of genius. Unless someone actually sits down with him and asks him directly and gets a direct response we will never know what his thinking is, but the conjecture using stats at least allows us to understand what he might be thinking.

    The comment from euroman was douche at best, but borderline racist. MOF is getting a run out because of the MEXICAN asst. coach. Really? And your support of this statement would be?

    After the last game we had a long discussion with mathewsf defending Beckerman. According to the passing stats, Beckerman was not the blackhole many had painted him to be. More than a few of us argued that he did not pass the eyeball test. Could Mr. Tuesday do a similiar analysis of him too?

    As always a pleasure to read and participate in such a fine forum on the greatest game in the world.

    Reply

    • I know Matt has done some work looking at the chalkboards for Beckermann. After this round of matches, we’ll see about working on another evaluation.

      Reply

    • Posted by Gregorio on 2011/11/09 at 10:25 AM

      Am I unamerican or a bad guy if I want MOF to mess up so we can call him MoFo?

      Reply

  26. Posted by 4now on 2011/11/08 at 10:52 PM

    This article has made me think about the conditions under which we watch football, and how important it is to have more options.

    I think that the provocative soft argument here is that you can’t really evaluate how a player impacts a system without being able to observe the whole system at work.

    This means that if you can’t actually watch the whole field at one time, then it is really impossible to evaluate how the system is working, and how elements are firing.

    Sadly, we, as spectators, very rarely get the chance to watch a full match. This would something really wonderful for networks to offer. The chance to watch a fixed camera from a high vantage point that showed the entire pitch.

    Without that, there is no way of assessing players or systems with any degree of credibility. Certainly not statistically, as well.

    Reply

    • Agree. The broadcasters don’t always make it easy to watch with attention on what’s happening both on, around and off the ball, but you can do it fairly well with an HD view.

      Please see the tactical pieces we did on the USA v CR and USA v Turkey before the worldcup. I got myself the ‘worst’ seat in the stadium to give readers exactly that viewpoint.

      Reply

  27. Posted by GeorgeCross on 2011/11/09 at 5:10 AM

    Interesting article Mr Tuesday.

    Are these stats as a result of playing the high line or playing Orozco-Fiscal? [I know that playing Oroxco-Fiscal gives you that option].

    We know the advantages and disadvantages of playing a high defensive line and the knock on effects this decision has with the set up of the rest of the team. I think it’s a great attempt to quantify speculation about what happened, although there are the disclaimers.

    If you think about the quality of the US team vs. quality of the CONCACAF opponents during WCQ, I think you’ll be witnessing a high defensive line, because against these teams, the US should be dominating possession, and this set up is better suited.

    Nevertheless, an enjoyable read. Appreciate the time and effort.

    Reply

    • Posted by scweeb on 2011/11/09 at 10:26 AM

      So what do we do after WCQ when are opponents get better? Should we still keep the high line or do we change?

      Reply

      • Posted by GeorgeCross on 2011/11/09 at 1:17 PM

        It isn’t binary, you can play high or deep and umpteen degrees inbetween. You play with one eye on your players and the other on the strength on the opponents, no?

        Reply

        • Posted by berniebernier on 2011/11/09 at 2:35 PM

          I think that is a great point being lost here. If its a tournament you might go with a deep line and say Gooch and Boca against a slow team that thrives on set pieces and then play Gooch and MOF the next game against a team that is quick forward where we want more possession.

          I applaud Klinsi for wanting to be able to play both ways (gooch and Boca played together last set of friendlies). Bradley seemed to play the same defense first and counter in all situations whether we were playing Argentina or Grenada.

          Reply

          • When we describe Bradley’s tactics, the standardized language is “defend (badly) and counter”.

            Reply

          • Posted by matthewsf on 2011/11/10 at 3:24 PM

            I know I belabor this point, but I like making it.

            The problem with Bradley’s team that led to poor defense is relying on that midfield to do too much and tasking 2 CMs with too much.

            Look at all the games, nearly every goal that wasn’t a simple execution/skill issue, came from the team being too stretched. The Mexico game, the Panama game, the Ghana game, the Slovenia game, the England. All of those games the midfield broke down somehow and the US had defend in small numbers.

            I went off on a tangent, but whether you like Klinsmann or not, the US is a) defending with 10 outfielders now, not 6. and b) defending further up the field.

            Reply

        • Posted by GeorgeCross on 2011/11/10 at 4:49 PM

          I haven’t had a chance to read all the comments, so I am not sure if this has been said before, but what are the possession stats? Looking at the number of passes can be a little mis-leading in isolation. If we’re talking about having possesion, then surely we should be looking at % of possession, and what area of the pitch said possesion took place, no?

          Also, identify the reasons for losing the ball / turn-overs would be interesting too[although, potentially difficult to quantify].

          Reply

  28. [...] professional punditry is that Michael Orozco Fiscal does not deserve to be selected by the USMNT. The Shin Guardian lays out why the pundits are [...]

    Reply

  29. Posted by fellaini's_fro on 2011/11/09 at 9:06 AM

    4Now. That is an excellent observation and suggestion. Much like high school football, there is a camera high above that films pretty much the whole field and you get a sense of what is happening during the play and off the ball.

    Considering professional sports is light years ahead in technology than high school football. I would have to guess the coaching staff has this type of film and are able to dissect the game off the ball. I would assume there are coaches high above as well that can give instant analysis as to what is going on during a game. Sadly as spectators our views are limited to our seats and television aspect.

    Reply

    • Posted by 4now on 2011/11/09 at 10:57 AM

      Thanks Fellaini. This reminds me of a time when I was working as an assistant coach for a college team. Another assistant, an older eccentric kind-of life-long first assistant had been ejected in the first minutes for protesting a call, and spent the rest of the match in the cheap seats. At halftime, he bounded into the locker room with all this vigor and revelation, “Holy crap! I can see everything from up there! I know exactly what is going on. Holy crap, holy crap!” he kept saying. It was strange, but reminded me how easily it is to get locked into a particular way of seeing and for this to become black-boxed, in a way.

      Reply

  30. Posted by Russell on 2011/11/09 at 10:51 AM

    Excellent article for the detail and thought provocation.

    I think MOF has not been terrible. And he allows a higher line than his current potential replacements. Therefore, Klinsman is hoping he’s learning on the job and getting better.. rather an alternative unknown that is likely not better. You see this type of decision making all over the place.

    Personally – I have seen MOF be in the right place at the right time to mop up some messes. Sure the CB should be there – but yes he was there.

    Can Tim Ream spend the off season in a marine boot camp? IF the kid could just toughen up and be nails.. he would be 5x better. The goals that i’ve seen that i hold him accountable for are ones that he did not take it to the offense. instead he got taken/caught watching.

    Reply

  31. [...] The Fan In You « Your Lying Eyes: What Klinsmann Sees In Orozco-Fiscal [...]

    Reply

  32. Posted by show me the data on 2011/11/09 at 6:07 PM

    To strengthen findings, perhaps we could see the same data for Orozco’s club team? If his impact on their performance is similar, then we may be on to something here…

    Reply

  33. Posted by @graysonson on 2011/11/10 at 4:40 PM

    Thank you for the factual, level-headed analysis. The whole of USMNT fans and pundits seem ignorant to the role of a ball playing, covering center-back. I don’t find Orozco Fiscal to be perfect, but he does offer some covering speed and awareness. Of the guys called into the team under Klinsy, I see our best CB partnership as Onyewu and Orozco Fiscal.

    Reply

  34. Posted by ztsuruda on 2011/11/10 at 7:08 PM

    I don’t completely understand what was proven in this article. I understand what the goal was to prove there are intangibles we don’t always see. However, your graph of statistics is a team effort. I could in general put any player as the variable and the statistics would generally be the same or very similar because its a graph showing team statistics.

    The graphs don’t show how long Orozco was on the field. I remember the US starting great and fizzling at the end of every game after Orozco was subbed out. Chip it up to Aguldelo, or Shea’s ran ragged, donovan not in, etc all reasons for poor possesion. Its exaggerated but maybe the team passes better because they know they can’t count on him. “don’t pass it to that guy, he’ll lose it”. You can’t pull him out and then say he’s the reason all these numbers look like facts.

    If Omar is so bad, why is he ranked so high in the MLS Castrol Index? wouldnt that be a statical method of assigning a number to field play? I’m not saying that’s correct either, but put Orozco through something similar with his club and then i’ll respect the numbers

    Reply

    • Posted by BernieBernier on 2011/11/11 at 9:03 AM

      The takeaway from the article is not that MOF is great or Omar is bad. Its that by having one faster CB then you can play a higher line. Playing a higher line helps with passing (and thereby keeping possession) and defending (as it makes it easier to press and more pressure should lead to better defense).

      The take away IMHO is that if Klinsman wants to play a high line (which has merits) then you don’t have 2 CBs you have a fast CB and a tackling/aerial/brut (we really need to come up with a term for this). If that is the case then Omar is competing against Gooch, Boca, Goodson for a 1 spot and not competing against Fiscal (since the positions require two different skillsets its almost like saying why doesn’t Omar play holding midfield as Omar was better than Beckerman in the Castrol Index).

      If you want to argue for a deep line and 2 tackling/aerial/brut CBs that is one thing (and then you probably get Omar over MOF in the roster) but that is a different argument – Klinsman should use different tactics – than Omar should be selected over MOF.

      Reply

  35. [...] only 12 appearances in the Clausura and – even if a small measure of tabular data suggests he was better than we all gave him credit for – he wasn’t exactly convincing in his national team appearances. Only 25, we likely [...]

    Reply

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Follow

Get every new post delivered to your Inbox.

Join 247 other followers

%d bloggers like this: