With all the raging debates about level of play in the league of late, It has gotten a few of us thinking about whether there are any stats that might reflect at all on the overall level of play in the league. There are a couple stats that have been produced so far that might shed light on the subject: 1 - The cumulative save percentage of the top 10 GKs over time has tended to rise over time. 2- The cumulative PK conversion rate (PKs on frame) is up over time, to near-'perfect' the last two seasons. Of these two, they are both sort of indirect indicators. I think they both might say something about what happens in the league, but both only cover a tiny corner of what happens. I wonder if there are other stats that might help as well. I thought this thread might be the place for suggestions, and I have a couple to start us off. -- One other measure I thought of to get at a piece of the league's quality in an indirect way is the percentages of assists and goals teams get from their top one or two players. My theory is a Carlos Valderrama or Carlos Ruiz wouldn't be (isn't) as dominant as in the past not because they are inferior (arguing Ruiz '04 is weaker than Ruiz '01 is an awkward point to try to make), but because defenses are becoming more and more able to constrain any team's main option, so that they must find more diversity in attack. If this is true, then these percentages would be going down over time. It would be interesting to see if that's true. Another is the entire league's shot conversion rate in Shots:SOG. I think this might say a little something, but especially in conjunction with number 1 above. The objection to the SP stat has been that it doesn't take into account what kind of shots were taken. My objection to that objection has been that it strains credulity a bit to believe the league's shooters, as a whole, were suddenly shooting more tamely over the course of an entire season. But if we had the SC rate we might see whether the argument was even more strained--theory being that if SP has gone up and SC stayed stable or gone up as well, then clearly keepers' SP isn't getting better because shooters were getting worse. The only two meanings I could draw would be that they were either both getting better coincidentally, or that the shooters had all shot more 'conservatively' as a strategic decision, not because they were just inaccurate. (And even on the latter explanation, if that decision were taken in absence of considering the defense, then surely players over time would have 'learned their mistake' and begun to shoot more aggressively again). I figure I might take some time and look at these numbers at some point, but I wanted to bounce them off the denizens here, so as to see if there are others to consider, or some reason I shouldn't consider these. So fire away!
As an addendum here, I will note that I am not seeking a definitive answer to the level of play debate--that's distinctly oout of reach, I think, since the stats themselves are almost all relative to the opposition, since MLS probably doesn't keep the best ones, since soccer is not the most amenable sport in the world to statistical analysis, and since some people will just never accept stats as a good argument. I just thought any additional info would be better than the usual shibboleths of the various factions that get tossed around on BS.
Good thread Stan. Particularly awkward because Ruiz 2001 was playing in Guatemala. I definitely agree with this point. What some people as a decline in talent, I see as general rising in quality. The same argument has been made about baseball - nobody hits .400 anymore, is it because hitters aren't as good as they used to be? Almost certainly not. Rather, it's probably because the gap between the outstanding players and everybody else has shrunk. Bad news, Stan. Like shot %, on goal % has been declining as well. I didn't remove PK's from these numbers, but I doubt they'd make much of a difference: first is shot %, second is on goal % Code: 1996 0.125 0.494 1997 0.121 0.486 1998 0.121 0.450 1999 0.104 0.433 2000 0.111 0.447 2001 0.115 0.468 2002 0.109 0.458 2003 0.110 0.447 2004 0.101 0.436 r= 0.831 Obviously, I don't think that players are becoming worse shooters (it would be nice if we had someone here who did). Rather, I suspect that improved defnese impacts both of these. I'd like to point out, however, that on goal % isn't really a measure of quality - some of the league's historically best scorers - Stern John, Damani Ralph, Edward Johnson, have pretty low on goal %'s.
Actually, it's the emergence of the strikeout that did the trick. It's damn near impossible to strike out 100 times and still hit .400, and 100 strikeouts in a season is actually better than average for a hitter.
So I've discovered a way to lure voros into the Stats and Analysis forum - post incorrect baseball information. I was just relying on my memory of an old Stephen Jay Gould book I'd read far before I had heard of Bill James for that claim. While you're here, voros, could you address my (apparently faulty) point - has baseball seen a leveling-off of talent since the Ruth era, or is the distribution the same as it's always been? Either way, do you have any recommendations for measures used to quantify this?
Chris, you'll forgive me for chopping up your quotes a little to make my responses more coherent: All of that is worth knowing, regardless. It would have made the case clearer, but oh well, it still adds to the conversation. As do I. But I will bring in your comments from the age thread, because it's relevant here: This allows the argument, which is part of what the more sophisticated among that camp has said, that the decline in both shot % and the rise in save % could be due to the same thing--poorer service that comes from younger, less experienced, and possibly lower quality players. And while the level of the lower 50% of MLS players may possibly have gone up a good bit even in the last 4 years, it could be construed that by replacing older, more expensive internationals with younger, less expensive, less experienced and possibly less talented Americans, you've also effectively 'creamed off' the top players over time, prudicing a race to the middle. I think that argument is probably not 100% wrong, in the sense that the young Amis don't seem to give the quality of service that the Valderramas and Etcheverrys used to (and Moreno and Guevara still do), even if it is also true that better organized, faster, and more athletic defenders also make that service more difficult to provide. However, to follow your next point. . . We have to ask ourselves "why would the league do this?" I can think of a few reasons: 1. To save money. The young Amis are cheaper. (We'd have to take into account also how often the league may have been burned in the past on high dollar imports at the position). 2. If the lower players are becoming better, then you can't expect #10s to just dominate the way they used to, and the money is better spend being 'spread around' among several offensive players. 3. They're doing it on purpose precisely because the long-term dependance on old, foreign AMs is bad for the development of MLS, and even more importantly for the NT. -- Now, usually I don't give the league credit for thinking farsightedly enough about the play to consider option 3, but it's possible. If it was all number 1, you could make the case that the league has been penny wise and pound foolish (though it could still accidentally have long-term benefit). On number 2, it would imply the league is not uniformly getting worse, but getting better at the median. And there you have an at least partially empirical question, which is why I should look into where the assists (and the goals, too, I think) are coming from at some point.
I don't think that this measurement is conclusive - people are just as liable to claim that it's caused by the good players leaving as the average ones getting better, but I like it anyways. Below are individual goal-scoring (non-PK) numbers, represented as a percent of total possible goals. So, e.g., the first entry under 2002 represents Carlos Ruiz's 21 nonPK goals, which represented 5.5% of the league total of 381. What do these numbers tell us? Well, first of all, it seems that the goal-scoring leader is very highly random, as percentage has a .05 correlation to year; complaints about the golden boot this year going for 12 goals may be attributable to mostly bad luck. 2004 looks to have been the year that we had the most parity among goal-scorers, as is pretty easy to observe by the fact that, after #8, 2004 leads throughout. This isn't a developing trend, however - before 2004, there wasn't much evidence that this kind of increased individual parity was emerging. Code: 1996 1997 1998 1999 2000 2001 2002 2003 2004 4.9 2.9 3.9 3.5 4.5 4.0 5.5 3.9 3.5 3.9 2.7 3.0 3.3 2.9 3.2 5.2 2.9 3.2 3.9 2.5 2.7 3.3 2.7 3.0 3.9 2.9 3.2 2.9 2.5 2.4 2.7 2.7 2.5 3.7 2.6 3.2 2.9 2.3 2.4 2.7 2.7 2.3 3.4 2.6 2.9 2.7 2.3 2.0 2.3 2.7 2.1 3.1 2.6 2.9 2.7 2.1 2.0 2.3 2.7 1.9 2.9 2.6 2.6 2.5 2.1 1.9 2.1 2.2 1.9 2.6 2.6 2.6 2.5 2.1 1.9 2.1 2.2 1.7 2.6 2.4 2.6 2.3 2.1 1.7 1.9 2.0 1.7 2.4 2.1 2.6 2.3 1.9 1.7 1.9 1.8 1.7 2.4 2.1 2.3 2.1 1.9 1.7 1.9 1.8 1.7 2.4 1.8 2.0 1.6 1.7 1.6 1.9 1.8 1.5 1.8 1.8 2.0 1.6 1.7 1.6 1.6 1.6 1.5 1.8 1.8 2.0 1.6 1.7 1.6 1.6 1.6 1.5 1.8 1.8 2.0 1.6 1.7 1.6 1.6 1.4 1.5 1.6 1.6 1.7 1.6 1.7 1.4 1.6 1.4 1.5 1.3 1.6 1.7 1.4 1.7 1.4 1.6 1.4 1.5 1.3 1.6 1.7 1.4 1.5 1.4 1.4 1.3 1.5 1.3 1.6 1.7 1.4 1.5 1.4 1.4 1.3 1.5 1.3 1.6 1.7 1.4 1.5 1.3 1.2 1.3 1.3 1.3 1.6 1.7 1.2 1.5 1.3 1.2 1.3 1.3 1.3 1.6 1.4 1.2 1.5 1.3 1.2 1.1 1.3 1.0 1.3 1.4 1.2 1.3 1.3 1.2 1.1 1.3 1.0 1.3 1.4 1.2 1.3 1.1 1.0 1.1 1.3 1.0 1.3 1.4
I just realized that I was doing the previous chart wrong - I need to use per team goal totals, rather than leaguewide. Revised list (numbers are arbitrary): Code: 1996 1997 1998 1999 2000 2001 2002 2003 2004 4.9 2.9 4.7 4.2 5.4 4.8 5.5 3.9 3.5 3.9 2.7 3.6 4.0 3.5 3.8 5.2 2.9 3.2 3.9 2.5 3.2 4.0 3.2 3.6 3.9 2.9 3.2 2.9 2.5 2.8 3.3 3.2 3.0 3.7 2.6 3.2 2.9 2.3 2.8 3.3 3.2 2.8 3.4 2.6 2.9 2.7 2.3 2.4 2.8 3.2 2.5 3.1 2.6 2.9 2.7 2.1 2.4 2.8 3.2 2.3 2.9 2.6 2.6 2.5 2.1 2.3 2.6 2.6 2.3 2.6 2.6 2.6 2.5 2.1 2.3 2.6 2.6 2.0 2.6 2.4 2.6 2.3 2.1 2.1 2.3 2.4 2.0 2.4 2.1 2.6 2.3 1.9 2.1 2.3 2.2 2.0 2.4 2.1 2.3 2.1 1.9 2.1 2.3 2.2 2.0 2.4 1.8 2.0 1.6 1.7 1.9 2.3 2.2 1.8 1.8 1.8 2.0 1.6 1.7 1.9 1.9 1.9 1.8 1.8 1.8 2.0 1.6 1.7 1.9 1.9 1.9 1.8 1.8 1.8 2.0 1.6 1.7 1.9 1.9 1.7 1.8 1.6 1.6 1.7 1.6 1.7 1.7 1.9 1.7 1.8 1.3 1.6 1.7 1.4 1.7 1.7 1.9 1.7 1.8 1.3 1.6 1.7 1.4 1.5 1.7 1.6 1.5 1.8 1.3 1.6 1.7 1.4 1.5 1.7 1.6 1.5 1.8 1.3 1.6 1.7 1.4 1.5 1.5 1.4 1.5 1.5 1.3 1.6 1.7 1.2 1.5 1.5 1.4 1.5 1.5 1.3 1.6 1.4 1.2 1.5 1.5 1.4 1.3 1.5 1.0 1.3 1.4 1.2 1.3 1.5 1.4 1.3 1.5 1.0 1.3 1.4 1.2 1.3 1.3 1.2 1.3 1.5 1.0 1.3 1.4 Using this list, there seems to have been a steady growth in the distribution of the scoring numbers, although I'm really not sure what kind of measures I should be using to analyze it.
Hm. I suppose you could do the means and standard deviations of each list separately. Code: 1996 1997 1998 1999 2000 2001 2002 2003 2004 54.00 47.70 54.50 58.30 57.90 55.00 57.90 51.60 55.40 2.16 1.91 2.18 2.33 2.32 2.20 2.32 2.06 2.22 0.98 0.44 0.77 0.87 0.96 0.82 1.26 0.64 0.66 That is the sum, mean, and standard deviation, respectively, of the columns. I think what the mean (and sum, really, since the mean is just dividing by 25) tells us is that the top 25 scorers have scored a relatively large portion of the goals on average compared to historically. I think the standard deviation is telling us that within that group, guys are grouped more closely to the mean, therefore there are fewer that stand out from each other. But I could be wrong on all that.
Wouldn't it be a good idea to contrast non-MLS leagues, where we have a good idea about the level of play, to see what type of statistical markers turn up? For instance, what are the statistical differences between the Permireship and Coca-Cola? Serie A and Belgium? If there are good stats for spotting league quality they should jump out there much more than between different years of MLS.
That's probably a better way to represent it, Stan. Yeah, that's what the standard deviation means - fewer high and fewer low values. However, I don't really see a trend here - .44 in 1997, 1.26 in 2002 - can we really make any judgments about which way the talent distribution is moving?
Well, comparing Serie A and Belgium presents a problem in that the playing styles are dramatically different, so the differences you see may be caused by that, rather than by differences in talent level. The same problem may exist in England, too, considering all the foreigners who play in the Premiership - I don't know. Otherwise, the biggest hindrance to me is simply finding the statistics - MLS keeps theirs tremendously well, as for other countries, it's really hit or miss.
'97 appears to be an outlier year, for some reason. It's not utterly convincing, but all the years other than '97 were more dispersed than these last 2. Both categories also spiked in 2002 (first season for Ruiz and Twellman?), but the standard deviation more than the mean.
I think it is a combination of all three: 1. Landon Donovan, DaMarcus Beasley, Eddie Gaven, Freddy Adu, Santino Quaranta, Kyle Beckerman, Bobby Convey, Danny Szetela, Ricardo Clark, Clint Dempsey, Kyle Martino, Ned Grabavoy, Memo Gonzalez, Justin Mapp, Craig Capano, etc. We are creating a nation of midfielders in the United States (with the exception being that we cannot seem to produce great right midfielders, a topic I've tackled elsewhere). Coaches realize that for most of their midfield needs today, they can invest in young Americans. 2. Give me the 96 version of Etcheverry or even the 98 version of Nowak, and I think they're very good MLS players, but not the world-beaters they were in their prime. And, since the US produces midfielders, if I were an MLS coach, I would be scouring the earth to find what the US hasn't produced enough of - outside defenders, right midfielders, and fast forwards. 3. I don't think MLS coaches are in a position to care about the long-term effects of their player selection decisions. They pick cheap foreignors who hopefully can help them win immediately. Some do it better than others, but at the end of the day, how well you perform in MLS comes down to how well your young American players perform for you.
Yes, all good, and all cheap, at least for a while. Here I differ. We seem to be short American #10s. I think this is more crucial, since wide mids seldom dominate games. As good as DMB was, he seldom took the team on his back the way Nowak often did. And I can think of exactly one American who can play a real #10 role at even a reasonably high level--Donovan (well, leaving Preki out, since he was first developed abroad and is now quite old). This is why I posited that maybe this turnover was a deliberate attempt to develop these players. I happen to believe that, too, but I'm at pains to try to support that belief. Agreed, but it depends on how much impact (veto power) you think the league central office still has. They could be blocking the contracts of older #10s deliberately in order to make sure Americans see the field at this position. It seems like generally, the group of Cienfuegos, Nowak, Etcheverry, and Valderrama were pretty successful. I can remember some of the busts like Branco, but I wonder if the failures were enough to prove this strategy wasn't a winner, considering nearly all the championships went to teams with older, more experienced, and usually foreign #10s. You would think teams would still want to try that strategy as much as ever, considering its track-record. (Even this year's DCU team became noticably better after Gomez got here. And before he got here, Moreno was playing less like a pure foreward and a whole lot like those guys I just listed.) They needn't always be young. Chicago won its championship with a lot of help from a younger Razov and a 'middle-aged' Armas. But you're right, you can't buy a title with foreigners and high priced Americans. You've got to have your Brian Namoffs and Josh Gros's, guys other teams underrated.
Sorry for the late reply. Like you could guess, this is a nearly impossible question to answer. The studies I have done seem to indicate that the players of the 1990s were simply better players en masse than the players that came before them. That's hardly earth shattering, but the stat work I've done seems to bear this out. The greats would still be greats. But the average player in 1936 or 1937 would likely have a very tough time in 2004 if you time traveled him. Some think it's better conditioning, and while that's part of it, the real difference is that the pool of potential players has exploded well beyond the extent that the league has expanded. Not only are there more Americans of every stripe, but blacks are now allowed to play and you have a huge portion of the majors consisting of foreign players, whereas they were oddities back then. In terms of dominating their respective leagues, yes the big kahunas back in the day dominated more than the guys recently, but only up until around the 1940s when that changed. From 1940 until about 1993 the difference between the best players and the average players was smaller than it was before and after those dates. A study of the 1943-1945 seasons (theoretically important seasons for any study of level of play since 75% of the league was away at war) didn't turn up any statistic you could point to and say "A-HA!" as the benchmark for level of play. Would does look apparent is that the 1943-1945 seasons, coupled with post-war integration significantly improved the level of play, particularly in the National League which had lagged a bit behind the American league. In most instances, expansion has ultimately increased the level of play, but interestingly I've yet to see that in the last two expansions. I don't think you're going to find the smoking gun in any one single statistic. I think the best you'll do is indicators. If you have detailed stats on all of the English leagues, that might be a good way to try and get at the issue. My guess is that on a very minor and far from absolute level, fewer goals is an indicator of a better quality league. It makes sense and what few numbers I've seen seem to suggest that as well. Maybe I'll pull my weight around here and list goals per game from the English leagues for the last several years.