[ChrisE]

Well, that's why I'm making this thread, at least. In the 8 years of the league's existence, the league has averaged .107 run of play goals/shot attempt. That number has seen a small decrease from the league's advent to now:

1996 0.116
1997 0.114
1998 0.116
1999 0.100
2000 0.104
2001 0.108
2002 0.102
2003 0.099

But it's remained pretty static. However, there's several guys who have, over their careers, almost doubled that number. Of players who have taken more than 80 shots in their career, these are the top 20 in career RoP goals/shots:

Savarese, Giovanni		0.228
Shannon, Musa		0.215
Pineda Chacon, Alex		0.214
Prampin, Alan		0.205
Marino, Pete		             0.202
Lassiter, Roy		0.199
John, Stern		             0.197
Twellman, Taylor		0.197
Valencia, Adolfo		0.191
Welton		             0.185
De Avila, Antonio		0.176
Graziani, Ariel		0.174
Donovan, Landon		0.173
Takawira, Vitalis		0.172
Wolff, Josh		             0.169
Ruiz, Carlos		             0.169
Cerritos, Ronald		0.167
Diaz Arce, Raul		0.165
Buddle, Edson		0.164
Washington, Dante		0.163

I guess the most notable thing here is that 6 of the top 7, and 9 out of the top 11, played primarily during MLS's early years (so to speak). I'm not really sure what to attribute this to (I never saw Alan Prampin play), perhaps an increased range in talent.

On the other hand, there's several guys who probably shouldn't be shooting. Here's the bottom 10 in career RoP goals/shots (80 shot minimum):

Mulrooney, Richard	0.025
Burns, Mike	             0.025
Hernandez, Daniel	0.039
Albright, Chris	0.042
Corrales, Ramiro	0.043
Franchino, Joe	0.045
Cloutier, Braeden	0.046
Elliott, Simon	0.049
Donadoni, Roberto	0.049
Vaca, Joselito	0.050

First of all, I think this may be an indicator of why Joselito Vaca hasn't been successful in the league. Whereas the other terrible shooters (save Albright) are defenders or defensive midfielders, Vaca is an attackind mid. In their careers, these 10 players have scored 50 run of play goals on 1206 shots, one more goal than Giovanni Savarese with only 991 more shots.

This year, eight players with 20 or more shots shot worse than 3%:

Cancela, Jose			0.000
Mulrooney, Richard			0.000
Hejduk, Frankie			0.000
Beckerman, Kyle			0.000
Garcia, Freddy			0.024
Corrales, Ramiro			0.026
Elliott, Simon			0.026
Guevara, Amado			0.029

Joselito managed to escape, ending up 17th this year, but you see three attacking, Latin midfielders who shot terrible percentages this year. Admittedly, Guevara scored some Open Cup goals that aren't accounted for, but it's an interesting trend. Also of note is that San Jose was either very very lucky to have Mulrooney make his shot in the Final, or Mulrooney takes very few shots like that.

This year's top 10 were:

Kovalenko, Dema		0.231
Noonan, Pat		             0.227
Donovan, Landon		0.212
Roberts, Zizi		0.185
Buddle, Edson		0.169
Twellman, Taylor		0.169
Simutenkov, Igor		0.163
Klein, Chris		             0.162
Kamler, Brian		0.162
Magee, Mike		             0.159

Twellman put almost exactly the same number as Noonan in his rookie year (.224), I'd expect Noonan to cool off a bit. Kamler's interesting simply because he never used to shoot before.

Here are some other renowned goal scorer's 2003 numbers:

Ralph, Damani		0.111
Spencer, John		0.132
Razov, Ante		             0.080
Mathis, Clint		0.099
Ruiz, Carlos		             0.105
Preki		             0.081

Obviously percentage isn't everything, but you have to wonder if Preki and Razov should be taking as many shots as they do. Ruiz's number fell of precipitously, from 0.219 to less than half that.



Also of note: Simon Elliott, after shooting a decent percentage in 1999 (.98), has now spent three straight years in the bottom ten.

Dema Kovalenko, who Karl Keller mentioned in a previous thread as perhaps this year's most efficient shooter, has seen a dramatic turnaround from last year. This year, Dema led the league at .23, last year he was 6th worst at .026. I have no explanation for this, hopefully some Fire or DC fans might offer some insight (in previous years, Dema was a good shooter, .159 and .196).

Chris Carrieri should score more goals next year; he saw a Dema-esque drop off to .056 from .203 (good for 4th in the league) and .25 in previous years. Some of this must simply be bad luck.

[kenntomasch]

One thing I do that I don't know is really that revealing, but is interesting to me, is to find out the percentage of shots that a guy puts on goal (Shots on Goal/Shots, minus PK's, of course) during the run of play. You can figure out what percentage of his shots on goal are saved (there will be a rare few that are blocked, but SOG-G will give you just about the number of times a guy had his shot saved by the keeper). You might be able to find out which guys get saved more often than others, which could be because they have a habit of shooting it right at the keeper, or because they played a disproportionate amount of great keepers or something.

So you could find out who has the highest percentage of his total shots saved and which guys are putting them off target. Like I said, I don't know how useful that would be, but it would be interesting to see.

[ChrisE]

Originally posted by kenntomasch
One thing I do that I don't know is really that revealing, but is interesting to me, is to find out the percentage of shots that a guy puts on goal (Shots on Goal/Shots, minus PK's, of course) during the run of play.

Yeah, I actually started out doing this, but it turns out that Simon Elliott gets a decent number of his shots on goal. He just never scores. Kerry Zavagnin is the worst in league history at SOG/shot. However, he still scores at twice the rate that Mulrooney or Burns does. Obviously, this may be luck, Zavagnin only has 5 goals in his career, but I'm not sure what this would show you that goals/sog don't. What do you think is the difference? I don't think that a shot on goal is in itself any better than a shot not on goal; an easily saved shot ain't any better than a shot that just misses the corner, or a glancing header that goes wide, etc.

You can figure out what percentage of his shots on goal are saved (there will be a rare few that are blocked, but SOG-G will give you just about the number of times a guy had his shot saved by the keeper). You might be able to find out which guys get saved more often than others, which could be because they have a habit of shooting it right at the keeper, or because they played a disproportionate amount of great keepers or something.

I think this is a good idea. I was thinking that maybe the reason Mulrooney, Elliott, Hernandez, etc. had such low percentages is that they were taking a lot of long shots. I'd imagine that a lot more long shots on goal are saved than close-in shots on goal.

So, basically, these numbers range from around .9 to .6, average of .76. The ten best are:

John, Stern		             0.522
Shannon, Musa		0.535
Pineda Chacon, Alex		0.604
Buddle, Edson		0.614
Savarese, Giovanni		0.614
Prampin, Alan		0.630
Valencia, Adolfo		0.638
Welton		             0.642
Lassiter, Roy		0.650
Serna, Diego		             0.650

And the ten worst are:

Burns, Mike		             0.938
Mulrooney, Richard		0.933
Albright, Chris		0.913
Corrales, Ramiro		0.898
Torres, Johnny		0.891
Cloutier, Braeden		0.886
Vaca, Joselito		0.884
Lewis, Eddie		             0.882
Hejduk, Frankie		0.881
Convey, Bobby		0.881

So you could find out who has the highest percentage of his total shots saved and which guys are putting them off target. Like I said, I don't know how useful that would be, but it would be interesting to see. [/B]

I guess it's interesting to see four of our current national teamer's and one former phenom on the most-saved list, but to me it looks like it's measuring basically the same thing as Goals/Shot on goal...

edit: another comment about Carrieri; if he had shot what he did last year, he would have ended the year with 11 run of play goals and 8 assists, putting him putting him at third in the league in run of play points, behind only Twellman and Preki. Obviously he didn't, but I'll be interested to see how he does next year.

[ChrisE]

Using positions listed on the MLSnet.com rosters, the RoP goals/RoP shots numbers for MLS's 8 years were (approximately):

Forward: 1929/14410, .134
Midfield: 1301/14479, .090
Defense: 418/4982, .084

I tried dividing these up into the categories Soccer America used to rank american players, Striker, Forward, Attacking Mid, Defensive Mid, Left Mid, Right Mid, Center back, and Wing back (I decided not to use right and left, since i have no idea where ritchie kotschau plays, etc.). I don't think the striker/forward distinction means much, but some other numbers are interesting:

Striker: 126/934 - .135
Forward: 78/637 - .122
Attacking Mid - .066
Defensive Mid - .06
Left Mid - .107
Right Mid - .085
Wingback - .058
Centerback - .064

The only really interesting things here are 1. that there's a big difference between the flank midfielder's conversion rates and the central midfielder's, and 2. the attacking midfielders are surprisingly similar to the defensive midfielders. I would have expected a significant difference between the two.

Regarding the defensive mid/attacking mid similarity, I checked their assist/90 numbers, wondering if they weren't in fact playing similar roles. However, no, attacking mids averaged .319 assists/90, defensive mids averaged .088/90. Attacking mids scored a good deal more (.090 goals/90 versus .051), but that's only because they took more shots.


[Assists/90 for other positions: strikers averaged .192, forwards .208, left mids .185, right mids .22, wingbacks .069, center backs .031.]

[mpruitt]

Great stuff. One thing that I'm going to try to work on is individualized lists of where a player is shooting from. This might have to start next season, hoping that ESPN Matchtracker will do MLS games. There was a discussion in the US Mens forum about American players not being able to shoot from outside the box. I referenced the Stats from the MLS playoff media guide, but unfortunately as a group there's nothing to compare them yet to, or at least I haven't taken the time to find similar information for the US Nats or other leagues.

[beineke]

Great stuff, Chris, simply top-notch.

A few comments:
(1) Calling them "Run-of-Play" goals is a little mis-leading, since goals from free kicks and corners are also included in your numbers, right?

(2) You mentioned that Mulrooney's goal came from a breakaway; it seems worthwhile to point out that Elliott's 2003 goal was also a breakaway that the opposing keeper played poorly.

(3) My favorite part of your work is the way you formalize the differences by position. But it raises the chicken-and-egg question: do forwards have better shooting percentages because they're playing forward, or are they playing forward because they're better shooters?

One way to address this would be to look at multi-position players -- does their shooting percentage change depending on what position they're playing? Unfortunately, the differences are small enough that I'm not sure we could get enough data to really tease out a difference.

[ChrisE]

Originally posted by maxim-1
Great stuff. One thing that I'm going to try to work on is individualized lists of where a player is shooting from. This might have to start next season, hoping that ESPN Matchtracker will do MLS games...

This would be a great thing to know. How would you be planning on identifying the shots, though? I don't think it's enough to simply identify the location of the shot. There's other factors, e.g. breakaway or not; whether or not it was off a cross, header/foot...

[ChrisE]

Originally posted by beineke

A few comments:
(1) Calling them "Run-of-Play" goals is a little mis-leading, since goals from free kicks and corners are also included in your numbers, right?

Yeah, my bad. If you, or anyone, has something else to call this kind of stat (other than non-PK goal), I'd be glad to hear it.

(2) You mentioned that Mulrooney's goal came from a breakaway; it seems worthwhile to point out that Elliott's 2003 goal was also a breakaway that the opposing keeper played poorly.

Good to know. I looked through the goals scored by the currently active worst career shooters this year (Mulrooney, corrales, franchino, elliott, vaca, pareja, andy williams, chris armas, bobby convey and jeff agoos), and tried to see if the goals had anything in common; obviously, this information isn't very useful without knowing how they typically shoot, nevertheless:

of 12 goals:

2 free kicks (Agoos, Franchino)
4 shots from outside the box (zavagnin, williams, convey, armas
3 crosses taken scored from pretty close (all w/feet)
1 amazing mazy run leading to a 1-on-1 (vaca)
1 mid-range/through traffic (williams
1 Mulrooneyesque unmarked/far post goal (armas)

Not much there, really, though i'd like to compare those to the goals by a high-scoring forward or Mark Chung.

My favorite part of your work is the way you formalize the differences by position. But it raises the chicken-and-egg question: do forwards have better shooting percentages because they're playing forward, or are they playing forward because they're better shooters?

One way to address this would be to look at multi-position players -- does their shooting percentage change depending on what position they're playing? Unfortunately, the differences are small enough that I'm not sure we could get enough data to really tease out a difference. [/B]

Good point, but, an additional problem is, how often is it that a guy plays two distinctly different positions? Even if you slot chris carrieri in at left mid or jason kreis at a-mid, they're not going to be playing like your 'average' player for that position would be playing.

[mpruitt]

Originally posted by ChrisE
This would be a great thing to know. How would you be planning on identifying the shots, though? I don't think it's enough to simply identify the location of the shot. There's other factors, e.g. breakaway or not; whether or not it was off a cross, header/foot...

I really think there might be some promise with the matchtracker stuff, as I believe they indicate what's what in regards to all that.

[beineke]

Originally posted by ChrisE

of 12 goals:

2 free kicks (Agoos, Franchino)
4 shots from outside the box (zavagnin, williams, convey, armas
3 crosses taken scored from pretty close (all w/feet)
1 amazing mazy run leading to a 1-on-1 (vaca)
1 mid-range/through traffic (williams
1 Mulrooneyesque unmarked/far post goal (armas)

Not much there, really, though i'd like to compare those to the goals by a high-scoring forward or Mark Chung.
[/B]

In Kenn's "goals" thread, there's a breakdown for how all of this year's regular season goals were scored. Only 44 of 353 came from outside the box. For this group, am I right in saying 7 of 14 came from outside the box (two fk's, the four you list, plus Elliott)? If so, it looks a lot like the low-pct shooters are often the guys who are most willing to strike from distance ... oddly enough, three of the best players among the low-percentage shooters (Agoos, Mulrooney, and Armas) all hit pay dirt during the playoffs.

BTW, I do think that there are quite a few guys who make positional switches. Certainly, several of the current Quakes have spent a fair bit of time in their MLS careers at more than one position (Agoos, Lagos, Corrales, Donovan, Mullan). Maybe we should see if Peter Hirdt is willing to dig into this one ... I'm particularly curious to know whether midfielders shoot better when playing their "weak" side, e.g. a right-footer like Ben Olsen who plays on the left.