Hey guys, in a thread in the Stats & Analysis forum, we were looking at what happens when a player misses a shot. There are basically two results that we can divine from MLS's stats, a saved shot on goal, and a miss not on goal. From these, it's pretty easy to make a ratio like (saved shots on goal)misses, not on goal). It turns out the all-time league average is around .67 (i.e., 2 shots saved for every 3 off target), and is a little bit higher for forwards. There are some guys who are well above that number (Digital Takawira, 1.21), and some who are well below (Kerry Zavagnin, .24). So, the reason I'm here is that Stern John's ratio happens to be extremely low, .38, placing him right between renowned scorers Peter Vermes and Mike Sorber. John is 9th from the bottom, and he's the only forward in the bottom 20 (the second forward is Eddie Johnson). So, obviously, it isn't necessarily a bad thing to not put a high number shots on goal, if the ones you get on goal are pretty likely to go in. I wasn't a fan of the league back when John was playing, so I don't really have any insight as to what about John made him unique in this way. I was hoping that the wizened Crew fans here wouldn't mind providing some suggestions as to how exactly Stern John managed to score so many goals with so few shots on goal, and why exactly he didn't put more on goal. So, thanks, I'd appreciate any input. For comparison's sake, here are the numbers for other significant scorers for the Crew; Edson Buddle also happens to be one of the bottom five or so forwards, all time; Marino is 3rd best: Code: [size=1] John, Stern 0.38 Buddle, Edson 0.45 Mcbride, Brian 0.57 Cunningham, Jeff 0.58 Washington, Dante 0.86 Marino, Pete 1.36 [/b][/size]
Well, Strikers thrive on goals so they try to get them whenever they can. I'm sure a lot of these off target ones are from distance or just half chances that a striker has to take. We should be happy that he gets into a position to have a shot so many times. I wish i had the stats for Henry in front of me, as i know he has a huge amout off target, but he also leads the premiership in goals scored.
That's actually a decent point, Chris. You should run a correlation between this stat and shots taken per minute and see if there's anything there.
Well, McBride always liked to pick corners. Usually, in those cases, the ball goes in or it goes wide. Also, what's missing in all this is - what is the player's overall shot/goal ratio? A particular player may get a low percentage of his overall shots on goal, but if he shoots the ball 10 times a game... Zavagnin has a low ratio, and few goals. But he's often shooting from 20, 25 yards. Pete Marino usually tried to dribble the ball into the net.
"Stinky Pete" did indeed usually seem to prefer walking the ball across the goal line. Strangest thing you ever saw. I'm not much on statitics (I specialize in uninformed shooting-off-of-the-mouth) but I would think that a more telling ratio would include ALL shots on goal (saved by keeper + goals scored). It would give you a better idea of either how accurate a shooter he is and/or (and this is the big one) how selective he is. The question posed by the original poster here is an interesting one: WHY did John have such a low ratio? Part of the answer lies in the fact that, at least as I recall it, 1) Stern was never a banger - he picked his spots and, if he didn't have the shot, he tried to get i to somebody who did and 2) MLS defenses were pretty unsophisticated and relied mostly on pummeling McBride into submission with two or three guys which let John pick an open spot for receiving and shooting the ball, often with a single touch.
I don't know if this enters into the discussion, but John scored in bunches, too. He had numerous multi-goal matches, including four or so hattricks. Maybe his goals-in-bunches has something to do with it.
I think voros was quoting Splarg, but talking to the thread starter, saying Splarg brought up a good point. Then again, my complete knowledge of voros consists of a few pages in a book, so I could be wrong.
First of all, I'd like to thank everyone for their responses. Captain Splarg, there is a correlation (about -.15) between shots/90 and this on goal/off 'accuracy' thing I'm talking about; that is, guys taking more shots show a small tendency to have more not be on goal than average. Stern John is 3rd all time in shots/90 among players with more than 2000 minutes played (4.39, behind Mamadou Diallo and Ante Razov); however, their ratios are both above average (.79, .78 respectively), and it's a pretty weak correlation, so I don't think this alone is enough to explain what the deal is with Stern John.
That's why I said that it's not necessarily a bad thing to not put a lot of balls on goal; Stern John's shooting percentage is 7th all-time among players with more than 80 shots(this thread has a lot about regular shooting percentages). What makes him particularly interesting is that he has such a good shooting percentage, and was such an effective scorer, despite not putting the ball on frame nearly as much as most of the league's best forwards.
Thanks for the input, Bill. You know I've had the same thought about shots on goal before, but I've got to (respsectfully) disagree. Stern John, during his tenure in MLS, had a 'on goal' percentage well below what is average for a forward (.42, vs. .50); however, of the shots he put on goal, 47% went in the net; that's the highest in league history, and is remarkably better than the league average of .27. Should we call him a mediocre-to-average shooter, because he didn't put enough on net? I'd say no. I think the problem with using sog/shots as a measure of accuracy is that the goal of a shot isn't really to put the ball on net, but to score. So hitting a ball off the post, or 2 feet wide, isn't really any less accurate than hitting the ball right at the keeper; they're both failures, but one gets recorded as 'accurate' and the other as 'inaccurate.' I don't really think it's possible to have one simple statistic that measures accuracy, which is why I'm trying to use this; it doesn't measure accuracy any better than does shooting percentage, but it provides an interesting (in my opinion) insight into what exactly a forward is doing, how he's scoring his goals. But if John's being very selective, wouldn't you expect him to put more balls on goal? I don't really follow the logic here.
I'd like to say again, I appreciate everyone's input, but I think I might not have been as clear as I should have been in the original post about what kind of a description I wanted. I tried to come up with some descriptors for forwards and how they scored goals, which (in my opinion) might influence these kinds of scoring rates: Did he shoot (in general) from close in, or medium distance, or from outside the box? Were his shots more powerful or controlled? Was he more of a poacher or a dynamic forward? Did he create his own shots or did he rely on service? Was he an aerial threat, or did he score mostly with his feet? Was he more effective in counter-attack or as part of a steady, build-up offense? Was he a selective shooter, or did he shoot whenever he could? Obviously, these aren't all the descriptive terms; some may not really apply, some may be meaningless, and I'm sure there's many that I've never thought of. So, if you feel like being really generous, answer whichever ones apply, or give me your own. Thanks a lot.
Statistical self-abuse aside, I would posit the apparently radical notion that ACTUAL GOALS SCORED might be a good indicator of so-called "accuracy". Stern was not too shabby at this part of the game. You know, putting it past the keeper, and all that. Like the night he put three past Amman. Or the night he put three past Presthus. Or the night he put three past Cassar. He seemed to have decent accuracy those times. Perhaps I'm just too simple for all this crap. Slot the fucking ball, win the match, grab some beers. This ain't baseball, with the Elias sports bureau spewing out the "on-base percentage of left-handed second basemen of Polish extraction with two strikes in the eighth inning with at least one runner on base in games their team was leading by three or more runs". Piss on all that.
I think your answer is pretty simple - he tried to pick corners. Stern didn't just smack the ball in the direction of the goal, he tried to place the shots where the GK can't get to 'em. Much like McBride usually did. If you aim for the middle of the goal, you can have less accuracy and still get the shot on-frame - but you probably won't score a boatload of goals. If you aim for a corner, and have good accuracy, you are probably going to miss the frame entirely more than shooter A - but you are also likely to score more goals.
I tried giving him the "John placed the ball rather than just flailing at it" argument, but he didn't like it. Apparently the perspective of having sat there actually watching a player for dozens of games rather than playing with the numbers five years later is insufficiently geekish.
Nicely said... this whole statistics thing is just absurd. If you want to be a stat-geek in determining who is a great player, then you're going to miss some. Statistically, how would you decide who is a good defender? There are very many good stats to determine what kind of a man-marker somebody is, or how good they are at positioning themselves and covering for teammates. What about keepers, because they are so dependant on the defense in front of them, their stats won't always be indicative of how good they are or how bad they are (in the case of a supreme defense in front of them).
I think that Foos (and Bill) are correct here. Its been a while, but I remember Stern making the chances for himself and then either dishing it off at the last second, or shooting a bullet that either scored, or went wide by the smallest amount. Either way, it was exciting to watch! What part of the word selective don't you understand? Selective means that he will not shoot "willy-nilly" towards the center of the net, but wait for a sure opening --- even if the opening is in one of the corners and might mean him missing the frame by a small margin.
I just loved to watch the Stern Turn. I remember being down in one of the corners when he scored one of those during a hat-trick performance. It was just beautiful. With his back to goal, the ball came in on a cross from the right side. He did a falling away shoulder trap with his right shoulder, continued his spin while the ball was falling, and then when he finished his 180, the ball was already in the back of the net. I think that was like my very first Crew game at CCS. I can't remember, but I think it was an afternoon game against Miami. Does that sound right to anybody else?
Yeah, Summer of 99 i think. Score were either 5-1 or 4-0. The next or (previous) game were against Dallas, who we also beat by 4 goals. I was very drunk for these two matches... those were the good old days.
Saturday, July 24th 1999 against Dallas. Sunday, August 1st against Miami. I was at the Miami game with friends. We still remember how freaking hot it was that day. Game temperatures were 88 degrees and 90 degrees, respectively.
Did you actually read my post, or did you just see the word 'statistics' and break out the old, tired 'statistics are worthless for the beautiful game' argument? I never said anything about ranking defenders using statistics - I recognize how difficult that would be. Furthermore, I never said anything about determining a forward's 'greatness' using statistics - in fact, I specifically said that whether they're high or low on this metric doesn't mean a whole lot about how good of a player they were. But, if you want to keep attacking this absurd straw man, far be it from me to stop you.
Gee golly, I'm really torn up now. I mean, what should I do? Should I use the "cry baby" smiley? or Should I use the "yawning" smiley? Yeah I read your posts, but it all sounds like rubbish to me. Have fun dorking out with your numbers. Hope it brings more enjoyment to soccer for you, but I'm not gonna get into an arguement over whether or not shooting accurancey is a relative function of a cosine curve.