Peter Hirdt presented some empirical evidence that the "Catch/Punch" statistic is really valuable ... it seems to correlate with winning.
The underlying logic is that a keeper who breaks up attacks will face fewer shots (or fewer dangerous shots). However, if you look at keepers from different teams, there isn't much evidence of this. That's natural, because a team that spends a lot of time defending will have a keeper who gets a lot of Catch/Punches and Shots Faced.
This is where Cassar and Garlick come in. They've both played behind the same defense and allowed similar numbers of goals per 90 minutes (Cassar 1.40, Garlick 1.32). However, they've come to those numbers by different routes ... Garlick appears to be a stay-on-his-line keeper, saving 4.05 shots per 90 while only getting 2.84 C-P's. Cassar appears to be much more aggressive, with 5.60 C-P's per 90. Perhaps this aggressiveness is the reason he only needs to make 3.13 saves.
Thoughts?
I don't know anything about statistics, but I do know the Burn fairly well. As you have suggested, numerista, Cassar is definetly more of a risk taker, aggressive type keeper then is Garlick. The team has done some shifting on the back, but for the most part both keepers have had similar back lines in front of them. However, Garlick got the majority of his starts in the early season, when the opposing team's offenses hadn't gotten up to speed yet. Cassar made some saves during his starts that might have been goals for many other keepers....possibly Garlick.....but he had to punch them out. There is no way to numerically prove this, but I believe had Garlick still been in the nets, his goal per average would be higher then Cassar's while his C/P statistics would be better. Just a thought. Some of those punches are saves that would have been goals. Broward County Florida Real Estate Directory and Islanders Vacation ::
Florida Real Estate Directory and Vacation Rentals. A directory of Florida real estate Agents, Appraisers, Brokers, D W V INVESTMENTS REALTY. DANGELO
http://www.floridarealestatedirectoryandvacationrentals.contyfloridarealestate.htmlHOME |
Elevated C-Reactive Protein in Patients With Obstructive Sleep Apnea ::
Imoberdorf R, Garlick PJ, McNurlan MA, et al. Enhanced synthesis of albumin and Pasceri V, Willerson JT, Yeh ET. A. Cassar, T. I. Morgenthaler, R. J.
http://circ.ahajournals.org/cgi/content/full/105/21/2462HOME |
Hi N,
I keep coming across this "fact" that c/p's correlate with winning & like you I'm pretty sceptical.
It tells you something about a keepers style,probably a little about his ability,but nothing about his team's ability to win.
Saves/shots on target tells you much,much more about a keepers ability...& even that doesn't guarantee to tell you everything about a team's win ability.
The crucial difference with c/p stats is that the keeper has a choice to stick on his line,let the defence deal with the cross & only make the play himself if he's confident of success.Compare this with a shot on target which he either saves or he doesn't.
A couple of figures taken from the Prem circa 2001.
Arsenal c/p for the 38 game season....64...finished 2nd.
Derby c/p.......154...one place off relegation.
Having figures for %age of catches dropped would help a bit also,the worst Prem keepers only drop on average 1 catch every 4 games. .
I've just had a quick look at the current league position/c/p stats in the mls,they're about as uncorrelated as you could get.
A case of any stat's better than no stat......not,...imo.
T.
No worries tachyon, the rep system has been tweeked again. It's now only your rep in the last 60-90 days or something along those lines. Frankly I think only a MetroStars fan could come up with something like this.
Thanks again for the article,Chris.I might well take you up on your offer of more when I've got my new PC sorted.
Hirdt's stats are certainly impressive,but it'd take a large leap of faith to say that more catch/punches cause more wins.
What's equally likely(if not more likely) is that the cause & effect is in the reverse direction.
Namely a team that's losing (& therefore more likely to lose) will eventually resort to more hopeful long range balls into the box,which could lead to the keeper of the leading team having more numerous & (possibly) easier high penalty area balls to deal with.
Hence more c/p for the keeper of the team in the lead.
A league of parity would probably extenuate this effect.
T.
Wow!
I mildly diss an esteemed stat & lose 76.9% of my rep points.
Lucky I didn't say it was certainly a case of wrong direction causation & therefore totally worthless & misleading......or I'd have lost the other 13.1% as well.
Next time I'll stick to less controversial topics like "why English goalkeepers are vastely superior in every department of the game to their American counterparts"
Sorta thought this was a DISCUSSION forum....I guess not.
T.
Sadly, T., this forum isn't nearly well frequented enough for any post to have a significant impact on your rep, anyway. :)
040619 Rapids v Dallas Burn Match Report::
Cassar attempted to punch the cross clear, but only redirected a long blooper facing injuries to goalkeeper Scott Garlick and forwards Bobby Rhine and Toni
http://www.bilfish.com/rapids/writing/040619RapidsDALmatch.shtmlHOME |
Here's a link to the pertinent thread: http://www.bigsoccer.com/forum/showthread.php?t=133224
Cheers Chris,
I've been trying to find the article since N posted,but the links kept coming up blank.
I'll get reading.
T.
Hi N,
I keep coming across this "fact" that c/p's correlate with winning & like you I'm pretty sceptical.
It tells you something about a keepers style,probably a little about his ability,but nothing about his team's ability to win.
Tachyon, I don't think you were around last year, I really doubt you were reading Peter Hirdt's column on mlsnet, so, in case you haven't seen it, here's his defense of the catch/punch:
No statistic in the MLS glossary draws as much disrespect as the “catch/punch”—a category that may be obscure to casual fans, listed under the heading “C/P” on the league’s goalkeeping ledger. Often criticized by traditionalists, sometimes to the point of ridicule, the catch/punch has become the Rodney Dangerfield of American soccer statistics.
The criticism is unwarranted. In fact, the statistic is such a good indicator of winning and losing that it’s foolish to dispute its validity as a measure of one important aspect of goalkeeping.
The catch/punch was designed to give goalkeepers credit for controlling their penalty box on attacks that don’t require saves. A keeper receives a catch/punch for catching a crossing pass, punching a cross out of danger, or beating an opponent to a through ball and clearing it. I think all would agree that the intent of such a statistic has merit.
The following figures demonstrate that the results have great value as well. These are the records of home teams since 2000 based on whether their keeper or their opponent’s keeper had more catches/punches. Only home-team records were considered in order to eliminate any bias by scoring crews in awarding catches/punches.
W-L-T Pct.
Home keeper has more C/P’s
133–24–37 .781
Home and road keepers are tied
52–31–19 .603
Home keeper has fewer C/P’s
103–101–50 .504
Not only does the home team have a much better chance of winning when its keeper has more catches/punches than its opponent, but the bigger the margin the better its record. When the home keeper had at least three more C/P’s than the opposing keeper, the home team’s record was a remarkable 58–5–11 (.858). I say that’s remarkable not only because there have been only five such losses over three-and-a-half seasons, but also because this is simply one statistic—totally independent of what else happened in the match.
Let me put it another way. It’s almost impossible to find statistics that correlate as well with winning and losing as a positive margin of three or more C/P’s does. Home teams whose keepers made at least three more saves than their opponent’s keeper since 2000 don’t have nearly as good as record: 43–15–16 (.689). Home teams that took at least three more shots than their opponents have a record of 162–92–56 (.613). Home teams with at least three more shots on goal have a record of 156–36–47 (.751). That’s still not close, even though goals are a subset of shots on goal. If we double the margin and consider only home teams with at least six more shots on goal than their opponents, the record is 66–10–15 (.807). We’re not there yet, but we’re finally getting close to the record of teams with margins of three or more C/P’s.
Still think the catch/punch is a random category, the product of American minds that compile and publish statistics only because they can? Random, meaningless statistics have no correlation to winning and losing. This one does, bigtime.
As a postscript, the only leaders list that places Tim Howard at the top of the MLS keepers for 2003 is a list of catches/punches per 90 minutes. Howard is the runaway leader (4.79), followed by Pat Onstad (4.15), Tom Presthus (4.11), and Adin Brown (3.89).
Wow!
I mildly diss an esteemed stat & lose 76.9% of my rep points.
Lucky I didn't say it was certainly a case of wrong direction causation & therefore totally worthless & misleading......or I'd have lost the other 13.1% as well.
Next time I'll stick to less controversial topics like "why English goalkeepers are vastely superior in every department of the game to their American counterparts"
Sorta thought this was a DISCUSSION forum....I guess not.
T.
I hope you are kidding. They are doing some work on the servers and many folks have lost a bunch of there rep points.
Cheers Chris,
I've been trying to find the article since N posted,but the links kept coming up blank.
I'll get reading.
T.
MLS redesigned their website and a lot of the old content has disappeared - I've got pretty much all the old Hirdt articles saved, so if you, or anyone else, wants them, I'd be happy to email them.
 How much does getting a small tattoo on your hip/stomach hurt?
 Do anyone else have an itchy anus? ?
|
Garlick v. Cassar