Average age of MLS players

[ChrisE]

I think I've probably posted this before, but here's a brief look at the difference in ages between American MLS players and foreign MLS players:


U.S. Foreign % US
1996 26.50 28.70 0.684
1997 26.74 29.01 0.693
1998 26.70 29.16 0.666
1999 26.64 29.17 0.686
2000 26.69 29.40 0.708
2001 26.63 28.99 0.696
2002 26.46 28.23 0.706
2003 26.11 28.33 0.727
2004 25.97 27.73 0.742


I classified foreign players as anybody who 'developed' (yes, not a very quantitative measure) outside of the U.S.

At its peak, the difference between the average age of a US player in MLS and that of a foreigner was 2.70 years; this year, the gap tied the low set in 2002, at 1.76 years. Obviously, it's still true, to some extent, that MLS is a retirement league for overseas players. However, it doesn't seem at all intuitive to me that this will persist - as MLS imports more players from CONCACAF, and the old Euros struggle to hold a spot here (see Powell, Daryl), it seems likely that the gap would at least continue to fall. Does anybody have any numbers on this difference between other leagues (numerista - Mexico?) - are significant gaps between foreigners and domestics commonplace?


A lot of the age difference between domestics and foreigners is accounted for by the very oldest players - had Preki played 1800 minutes this year, the drop in foreign ages would have been half, falling from 28.33 to just 28.03. Additionally, simply removing the ten oldest players in the league removes a full year .6 years from the gap (dropping it to 1.16). That's because, while foreigners only account for 28% of MLS players (and only 26% of minutes), 6 of the 10 oldest players (and 7 if you count Robin Fraser [8 with that Swissman Agoos]) are foreign-raised. That looks likely to change soon, however, as Preki, Pareja, and Stewart can't hold out much longer, and recent imports Herzog and Hong are retiring after this season. (Old imports don't seem to be lasting as long as they used to.)

The oldest:
Preki
Fraser, Robin
Onstad, Pat
Agoos, Jeff
Pareja, Oscar
Herzog, Andreas
Myung-Bo, Hong
Meola, Tony
Stewart, Earnie
Jones, Cobi

[numerista]

Here are the minutes-weighted ages from Mexico, using age at (roughly) the start of each temporada. Without the weighting, Mexicans appear about a year younger, while the foreigners' ages don't change significantly.

My impression is that in Mexico, teams tend to have older domestic players than in most countries. They generally keep large squads, which makes it hard for youngsters to break in.

Mex. Foreign
1996-1 26.39 27.61
1996-2 26.60 27.86
1997-1 26.77 28.04
1997-2 27.07 28.11
1998-1 26.84 27.47
1998-2 27.00 27.45
1999-1 26.91 27.16
1999-2 26.97 27.59
2000-1 26.73 27.55
2000-2 27.19 28.20
2001-1 26.89 28.30
2001-2 26.71 28.66
2002-1 26.55 28.83
2002-2 26.63 27.72
2003-1 26.82 27.69
2003-2 26.82 27.89

[Stan Collins]

I would imagine the league will get older next season, as expansion prolongs careers, just as it got younger when contraction forced teams to choose between older players who were a little better right now, but got paid more and were headed downhill, and younger kids with an upside.

[numerista]

I would imagine the league will get older next season, as expansion prolongs careers...

In general, I agree with this logic. Players decline more slowly than they improve, so if you're going to extend a player's career, you'd rather take the guy who's a year too old, not a year too young. In addition, it isn't always easy to lure young talent away from college.

In this specific case, though, I'm not so confident about what will happen. There are several reasons why the expansion teams might themselves be on the young side...
History -- In the 1998 expansion, the Fusion drew heavily upon veterans. The Fire dug up younger players, like Thornton, Brown, Armas, Marsch, Razov, and Wolff. We know how that worked out.

Coaches -- With Rongen and Ellinger coming from the youth national teams, they should be closely attuned to talent at that end of the spectrum.
As an example, Santino Quaranta will be only 20 years old next season, and he has played for both of these guys.

Spanish-speakers -- To the extent that Chivas wants Spanish-speaking players, they'll be forced to rely on youth. Sure, they can have three senior internationals, but top Mexican veterans are none too keen on coming here. Ramón Ramirez is already making noises about backing out.
Instead, foreign reinforcements are likely to be up-and-comers -- they've been talking about signing six transitional internationals. In addition, hispanic players in the US tend to be younger, so that could help bring down the age, too.

[swedcrip34]

I would imagine the league will get older next season, as expansion prolongs careers, just as it got younger when contraction forced teams to choose between older players who were a little better right now, but got paid more and were headed downhill, and younger kids with an upside.

disagree, if MLS adds 4 developmental spots per team, then I think it'll go down

[Stan Collins]

In this specific case, though, I'm not so confident about what will happen. There are several reasons why the expansion teams might themselves be on the young side... Interesting points. Some of them make some sense. I personally don't think Rongen has tended to be infautated with youth in the past, and we really don't know about Ellinger until he coaches adults.

The Spanish-speaking thing is intriguing. It depends on how much Chivas is willing to risk sucking in their first season. The hiring of Rongen might indicate they think it's more important to get off to a competitive start than to stick too literally to the Spanish language (I personally think they took him ofver a Spanish speaking coach mostly because they needed someone with enough knowledge of the players in this league to get a good expansion draft). But then who knows.

swedcrip34,

You're right, I forgot about the developmental slots. But remember, we are also talking in terms of weighted by minutes played, so those guys might not play for the first team very much.

[numerista]

Interesting points. Some of them make some sense. I personally don't think Rongen has tended to be infautated with youth in the past...

He may not be infatuated, but neither did he hesitate to bring Convey and Quaranta along quickly. Overall, his track record in MLS drafts is hit-or-miss. In year one of MLS, Hejduk and Ralston were steals, but in subsequent years, he had some massive washouts like Johnny Torres and Jason Moore. His most recent draft haul -- Lisi, Nelsen, Quaranta, Namoff, and Ziadie in 2001 -- was excellent.

The hiring of Rongen might indicate they think it's more important to get off to a competitive start than to stick too literally to the Spanish language

I think that's correct, but it doesn't mean they'll abandon the Spanish-speaking philosophy altogether. They're going to look long and hard at the Raúl Palomares and Jesús Ochoa types, to decide who's ready to contribute.

In addition, Transitional Internationals could serve both purposes at once. I pulled up a recent squad list for Mexico's Under-20's, and six of its players are from Chivas. Those youngsters would all be options for Rongen, as would Isaac Romo, who played for their 2003 U-20's but is still stuck on the bench in Guadalajara.

[ChrisE]

History -- In the 1998 expansion, the Fusion drew heavily upon veterans. The Fire dug up younger players, like Thornton, Brown, Armas, Marsch, Razov, and Wolff. We know how that worked out.

This is actually not really true, n. Maybe people will remember it that way, and that will affect how the draft shakes out, but the Fire were actually an older team in 1998 than the Fusion (28.4 to 27.8) - the Fire's veterans - Nowak, Kosecki, Podbrozny, Kubik, Klopas, Okaroh, were all well over thirty and drove the average up significantly. Fire median age was far lower - 25.6 versus 27.5, because of an enormous gap that existed between the vets and the rookies. The 4 players between 25.6 and 31.3 played 908, 159, 11, and 6 minutes.



Coaches -- With Rongen and Ellinger coming from the youth national teams, they should be closely attuned to talent at that end of the spectrum.
As an example, Santino Quaranta will be only 20 years old next season, and he has played for both of these guys.

I certainly think that this is a significant difference between the last expansion draft and this one. With the developmental rosters and P-40's, there are going to be a bunch more players in the 21-24 age range that are appealing prospects - considering the # of allocation and discovery players expansion teams will have to fit under salary cap, I think we may see mostly this kind of player taken in the draft.



Spanish-speakers -- To the extent that Chivas wants Spanish-speaking players, they'll be forced to rely on youth. Sure, they can have three senior internationals, but top Mexican veterans are none too keen on coming here. Ramón Ramirez is already making noises about backing out.
Instead, foreign reinforcements are likely to be up-and-comers -- they've been talking about signing six transitional internationals. In addition, hispanic players in the US tend to be younger, so that could help bring down the age, too.

Although I doubt it will affect much of anything next year, the likelihood that most Chivas USA players will join the team without attending college as teenagers could have a long-term effect on driving the age down.

[ChrisE]

As a point of reference, here are the 1997 and 1998 ages for MLS clubs:


Age 1997 1998
Colorado 27.2 26.7
Columbus 27.3 27.8
Dallas 27.2 28.9
D.C. United 27.7 26.7
Kansas City 27.4 28.2
Los Angeles 26.8 26.5
Metrostars 27.1 27.0
New England 27.3 27.2
San Jose 28.6 27.4
Tampa Bay 27.8 27.9
Chicago 28.4
Miami 27.8
Average 27.4 27.5

Expansion 28.1
Non-Expansion 27.4 27.4

[numerista]

This is actually not really true, n ... Fire median age was far lower - 25.6 versus 27.5, because of an enormous gap that existed between the vets and the rookies. The 4 players between 25.6 and 31.3 played 908, 159, 11, and 6 minutes.

Interesting stuff again, Chris, although I don't really think this contradicts my point. The Fire did have some older veterans, but as the median age shows, the Fusion relied on more veterans. (Perhaps this goes back to the ambiguity of the word "veteran.")

Since this is the stats forum, I might as well ask: is this a weighted median?

[ChrisE]

Interesting stuff again, Chris, although I don't really think this contradicts my point. The Fire did have some older veterans, but as the median age shows, the Fusion relied on more veterans. (Perhaps this goes back to the ambiguity of the word "veteran.")

Well, it doesn't really contradict it, but I think it shows that we can't simply call the Fire a 'young team' and the Fusion an 'old team.' I think actually the best division we can draw about the two teams is one of salary allocation - although this is a guess, it looks like the Fire were bulit around a core of high-paid foreign veterans, with talented, but cheap, young American players filling in everywhere else. (I wouldn't be too confident that would work again.)

The Fusion, meanwhile, were more diffuse - they had one young, presumably highly paid player in Diego Serna, their draft picks Cullen and Mastroeni and P-40 Carlos Parra played a lot, while pretty much everybody else was an expansion draft/discovery pickup.



Since this is the stats forum, I might as well ask: is this a weighted median?

Yeah, sorry, minutes-weighted median. The non-weighted medians were 26.3 Fusion, 25.6 Fire.

[ChrisE]

Out of curiosity, I tried to figure out where the various minutes-played for both expansion teams came from. I divided players into 5 ways of acquisition - Expansion Draft, College Draft, Supplemental Draft, Allocation/Discovery, or P-40, and treated players who were acquired with players from one source as being from that source (so Chris Armas and Jorge Campos were "expansion" acquisitions). I wouldn't say that the results came out too meaningfully, both teams look approximately the same, but it took me long enough that I'm gonna post it:



Chicago Mimai
E 0.42 0.47
C 0.13 0.14
S 0.08 0.00
A 0.35 0.34
P-40 0.02 0.04


Although conditions are radically different this year, I'd expect to see something similar from RSL and Chivas.

[numerista]

It looks like the Fire were bulit around a core of high-paid foreign veterans, with talented, but cheap, young American players filling in everywhere else.

Would this really distinguish them from the Fusion, whose foreigners included Serna, Valderrama, and former Argentinian league star Marcelo Herrera?

At that point in time, MLS was dominated by foreigners. Nine of the twelve teams were led in scoring by a player who was raised outside the US. The exceptions the Burn (Kreis), the Galaxy (Cobi), and the Fire (Razov).

[ChrisE]

Would this really distinguish them from the Fusion, whose foreigners included Serna, Valderrama, and former Argentinian league star Marcelo Herrera?

At that point in time, MLS was dominated by foreigners. Nine of the twelve teams were led in scoring by a player who was raised outside the US. The exceptions the Burn (Kreis), the Galaxy (Cobi), and the Fire (Razov).

The chart below illustrates individual players' minutes played and ages.


1542 36.6 2294 34.6
1260 34.7 2790 34.2
1176 34.6 1921 33.7
1441 31.5 1497 32.2
1690 31.3 1999 32.1
2488 31.2 523 32.0
1784 29.6 765 31.5
1187 29.3 1979 31.3
90 28.9 6 30.0
1575 28.8 11 26.8
1263 28.4 159 26.6
212 27.5 908 26.2
524 27.0 2790 25.6
270 26.9 625 25.5
270 26.4 812 25.4
2232 26.2 2076 24.5
310 25.0 2199 24.4
1988 24.5 2091 24.1
1890 24.2 623 23.7
719 23.7 90 23.6
332 23.4 218 23.1
288 22.7 2470 22.8
323 22.5 1973 22.4
390 22.3 137 22.3
2680 22.1 651 21.1
1611 21.6
57 21.5
1317 21.2
90 21.1
360 20.6


On the left is the Fusion. On the right is the Fire. While they were both made up of a good number of foreigners, it looks to me like they're very differently constituted.

[mtr8967]

I'm not sure averaging is the right thing to do here. That will produce the same result for a pair of 27 yr olds as for a 34 yr old paired with a 20 yr old.

Eyeballing the data, I don't see any major difference between the two teams. Maybe it just seems that way because the Fire's young players turned out to be better.

[numerista]

Maybe it just seems that way because the Fire's young players turned out to be better.

I don't think that's true. Perhaps a clearer comparison can be obtained by looking at the acquisitions who came from the expansion draft. These are the players the Fire got from their picks ...
63: #7 Okaroh (SI)
64
65:
66: #4 Klopas
67:
68:
69: #2 J. Campos
70:
71: #3 M. Lagos, #10 Bates
72: #1 C. Armas, #5 Salcedo, #8 D. Gutierrez
73: #6 Thornton, #11 Marsch
74: #9 A. Lewis

Apart from internationals, they made sure their acquisitions were born from 1971-74, old enough for high productivity but young enough to be under-rated. They kept an eye out for players with low minutes, either due to injury history (Lagos, Gutierrez) or playing behind top talent (Marsch, Thornton).

The Fusion (below) took the opposite approach. Their top three picks were born in the 1960's and had no real prospects of ever earning a cap. Throughout the process, they seemed intent on players who had already been getting playing time. That's true even of the youngsters they chose ... with the exception of Cassar and the 12-pick Taylor, everyone had at least 1000 minutes under their belts.

63: #7 Kooiman
64:
65:
66: #1 Vaudreuil
67: #11 W.Webber
68: #3 Kelderman
69: #2 Maessner
70:
71: #10 Budnick
72: #9 Kmosko
73:
74: #5 Cassar, #6 N. Vargas
75: #4 J. Martinez, #12 B. Taylor
76:
77: #8 Corrales

The Fire realized they would find more talent beneath unturned stones.

[ChrisE]

Once again trying to return this subject to ages...

For another topic, I divided MLS players into positional categories - goalkeepers, forwards, offensive midfielders, defensive midfielders, flank midfielders, center backs, and fullbacks. I thought it might be interesting to compare their ages, in particular to look at differences among positions that MLS groups together (i.e., F, M, and D).

Below are the minutes weighted numbers for 2004:


GK 29.81

F 25.88

AM 24.86
DM 27.17
FM 27.10

CB 27.21
FB 26.08


The immediate surprise here is the attacking midfield position - what used to be the domain of MLS's foreign veterans, is now its youngest position. Much of this is caused by two outliers, one Freddy Adu and one Eddie Gaven, but even excluding them you get 26.3, well below the other midfield numbers. Even more interesting, this number looks likely to fall; last year, there were only three attacking midfielders in the league over 30 - Ronnie Ekelund, Andy Herzog, and Preki. Herzog has retired, Ekelund's contract wasn't renewed, and Preki's at least considering hanging them up.

I think these numbers support a claim that I'm not particularly fond of, that the league has lost the "creative #10" who can dictate an offense, etc. etc. But I also think it indicates that it's too simplistic to look at this as a decline in quality - I think it indicates a shift in who MLS looks to for creative play. The average of #10's in the league isn't likely to fall simply because the old players are retiring, but also because there are a host of young guys who are likely to get far more minutes next year - Adu, most obviously, but also Ned Grabavoy, Jordan Cila, Craig Capano, Andy Dorman, Ramon Nunez, and perhaps, Memo Gonzalez. Although the claim still stands that MLS hasn't produced good attacking midfielders - I think the emergence of Dempsey, Gaven, and Martino, as well as the players listed above, may indicate that this is about to change.

The other surprise to me was the fullback/center back difference - fullbacks are more than a full year younger than center backs, on average. It's not really possible to come up with a list of talented young fullbacks like you can talented young attacking midfielders - the youngest was 21.8. Two factors seem to have caused this: 1. developmental rosters - Bryan Namoff, Chris Leitch, Marshall Leonard, Philip Salyer, Denny Clanton, and Tim Regan all play this position; it's probably the position where developmental players have made the biggest impact (and, I'd infer, therefore the weakest).

2. The league doesn't have any old fullbacks - the oldest is Vuk Rasovic, who only played 45 minutes this year, followed by Milton Reyes (30.2) and Frankie Hejduk (29.9), and Craig Waibel (28.9). Whether this is because the position requires players with greater stamina (likely), or because of some historical deficiency in the position, or some other reason it's too late for me to come up with, I don't know. But it was certainly quite a surprise to me.

[scaryice]

http://usasoccer.blogspot.com/2009/03/mls-team-ages.html

Bumping this old thread to say that I've figured out the average age of every MLS team in history.

[soccer_fan123]

not that many young players...