PSP is very pleased to welcome back Dan Stover and Jim Prestifilippo, who present their infographic of Philadelphia Union stats for April’s games. Click here for a downloadable version of the infographic.
PSP is very pleased to welcome back Dan Stover and Jim Prestifilippo, who present their infographic of Philadelphia Union stats for April’s games. Click here for a downloadable version of the infographic.
Excellent job as always, Jim and Dan. This was quicker than I remember the last one coming out, so bravo. Some good points can be gathered from this, especially with the larger pool of data.
.
Keon is up at #2 on positive possession this month, and when factored for minutes he ends up just barely behind the CBs and D-Mids? Wow, looks like everyone who suggested that Keon loses possession was sorely mistaken, as that is way off base. In fact, the converse is true. Hmm 😛
.
LeToux with a -4 while on the field this month…ouch. I think it’s probably a safe bet that we’ve seen the last of the Frenchman. That is fine by me, by the numbers I would rather see Gabe or even Lahoud come on late rather than LeToux. He is subbed on to press for goals, but it’s interesting that the opposite happens.
.
Mike Farfan has fairly poor passing %’s and even worse possession. Worse than Cruz, which is surprising. Maybe the fact that his defensive numbers are so high are because he has to recover all of the balls that he gives up? Well, all the balls that him, Sheanon(this hurts to say), and the forwards(to be expected) turn over.
.
Side note for Dan and Jim: I really like the addition of the Keys to show the values of some of the objects(in possession and defensive plays) as well as the subtle divider between the month on the left and year on the right.
– C-Dawg
Marfan has a -4 for the season. Should we see the last of him as well? The +/- is basically meaningless to me…case in point: Jack is rated at 0 – ‘nuf said.
.
Not the last of Marfan yet, as his last performance was surely much better than all the rest. However, how can the +/- be meaningful? I am a big hockey fan as well, and I always think +/- is very telling. I understand the difference between the two sports, shifts in hockey vs being on the field constantly for soccer, but I still think it is just as telling. If mike was on the field for 4 more opposition goals than Union goals, could it be possible that he was partially at fault for any of them. Or could it be that when he is on the field, the Union are not scoring enough due to a lack of possession and passing? While you simply dismiss this stat, I think it is a valid conversation piece. It doesn’t tell the entire story, but it does add something to think about.
.
BTW, the LeToux comment was based on the worst +/-, his awful possession numbers, his awful passing numbers, and lack of offensive contribution. So yes, the stats back up LeToux riding the pine.
.
To take your JackMac example under a microscope:
When you look at the team averages for +/-, they are mostly coming out around -2 or -1. JackMac is a nice even 0. That means that when he is on the field, the team is performing(outscoring opponents) better than when Jack is not on the field. When looking at Mike’s +/-, the team is performing twice as poorly as when he is not. And when LeToux is on the field, the team is performing 3x as poorly and so on. Could it be that bad possession numbers and bad passing numbers leads to less goals for and more goals against? Unspeakable…
Parke, Okugo, Williams, and Carroll have played full shifts all year. As a team we have an aggregate -2 differential. C-Dawg is correct in the assumption that when Jack Mac is on the field we maintaina an even score line, if not win games. He is also correct in stating that when Marfan or Le Toux are on the pitch we are conceding. Now, the reasons for that might be different and are essentially game-by-game situational. To make a blanket statement that the stat is meaningless is off-base becauase it has the ability to show which players [especially bench players] are playing effectively for a positive result. I think, with more time, the stat will balance out and possibly provide a clearer picture come November.
Wow. Always enlightening and full of surprises. Even though I’m not a Le2 fan I am surprised at how bad he looks in this.
This article exposes Williams…….
It’s painful to see it shown like that. Hopefully he can turn it around. I think there is more there than just Williams though. I think a lot of it comes from Cruz being there. If he had more of a complete midfielder there I think the numbers would look a lot different. Maybe the Union’s record and point total would look different as well. Who knows..:[
I chalk a lot of Williams’ “poor” showing to his inability to connect and play off of Cruz. He is making a ton of passes but having difficulty reaching teammates. I think if Garfan or Marfan started wide right, his numbers would be better.
You hit this one perfectly. Cruz is so eager to try and get behind the defense that he leaves Williams with no easy outlet for the ball.
Always an interesting read. The most meaningful stat for me is offensive productivity for 90 minutes. Nothing is more important than goals, right?
.
Is there a way to add “missed mark” on a goal to the defensive ratings? This is a crucial piece in evaluating defensive performance.
.
One other suggestion – it would be nice to see some keeper stats. Maybe compare Zac’s save percentages vs. other MLS tenders? Not sure.
The Offensive Prod. per 90 also factors for assists. I have considered a “missed mark” kind of stat, but it can become very subjective depending on the run of play. I felt that Dan, Eli, and the other guys who write for PSP do a strong game by game analysis, pointing out which players were at fault for a goal. In regards to MacMath and ‘keeper stats, I chose early on to omit them mostly because they are already tracked on MLS.com. Our attempt here is to track and present statistics that are not typically tracked in the hopes of 1. providing further analysis of the game, 2. generating conversation, and 3. give the PSP readers something to enjoy.
Good stuff! Thanks for compiling the data.
The Shots on Frame section is wonky. I understand it goes by percentage, but I’m unhappy with how it’s presented. BC went 0/3 and he’s listed as the worst and Hoppenot only shot four times and he’s listed as the best. I mean yeah, the percentages don’t lie, but BC missing his three chances isn’t as bad as Keon missing EIGHT of his twelve shots. I just feel like shot percentage is not the same as say, “shot efficiency” or something.
Exact same comments about the Passing percentage section. I get it, percentage is the key here, but you can’t be comparing guys who have passed 100+ less balls! For example, you’re comparing Kleberson’s passing percentage of 71 (he only attempted 17!) with Williams’ 64% (he attempted 408!!!).
I’m sorry, but that does not make for a compelling infograph. To be completely honest, I really didn’t give the rest of the infograph the time of day after I saw that first section. Come on, guys.
and If we represented the data by just total shots off frame or most passes completed, someone else would be echoe your response in regards to the percentage differences. So in order to please everyone, we included the percentages and figures so the reader could have all the figures to generate their own analysis.
That’s fair. You guys asked for input so I gave input. All in all I’m just glad somebody has recorded stats available.
Don’t get me wrong, I appreciate the input! Do you think maybe factoring for shot’s on frame per 90 minutes and successful passing percentage per 90 minutes would help/interest you?
.
Or something along those lines?
Yikes! It’s a good thing they’re the ones doing the infographic. Couple of points here:
.
I agree that it is somewhat odd that Hoppenot is at the top. However, he is up there because he is not wasting many shots. If I can tell correctly from the data, the players seem to be organized first by percentage and then I guess if there is a tie, they are ranked by number of total chances (for shots on frame). To me, this makes sense. For example, LeToux and Okugo both had 0% on frame this month, however Okugo is listed as worse because he has squandered more chances.
.
Percentage matters more than total chances because it evens the playing field. Same goes for passing. If you put the players with the most successes at the top, Sheanon Williams would be the top passer. That is clearly not the case, as he has close to the worst completion rate. They did exclude Torres and Albright, as the infographic says, for a low sample size. Based on that, I’m guessing 10 or maybe even 15 is the cut off for passing. What would you recommend as a cut-off point for this. How many passes is enough passes to really get a feel for who the most successful passer is? 10? 50? 1000?
.
IMO, anywhere between 50 and 100 is fine. If a player is continually making unsuccessful passes, then it isn’t just happenstance. It’s either them or their chemistry with teammates causing that. I think that you’ll probably find that as the season goes on, the year side of the infographic’s passing % section will grow stronger, while the month offers a look at who was on their mark the past few weeks, which is helpful in it’s own way.
.
Last comment: On what earth is 0/3 shots on goal better than 4/12? I would rather have someone with 1/100 than that same player having 0/anything. How would anyone with a 0% for shooting ever score? If you can’t hit the net you can’t score (not counting own goals of course but those are infrequent).
.
While I agree to an extent that the minimum sample size should be limited for certain stats(i.e. no less than 50-100 passes to be included in passing % category), there is no way that the data/metrics used to come to the final numbers should be influenced by total anything. Is the guy that goes 45/50 on passes worse than the guy that goes 400/500? No, and maybe the the guy with 40 completed deserves some extra minutes to prove whether or not he is better than the guy with 400. I think it’s sad for you to dismiss this simply because you don’t agree with the sample sizes. The guys who did this can’t make up numbers just so the sample sizes are even, and I think they do a pretty solid job of presenting it clearly and providing totals for certain things as well.
.
– C-Dawg
I actually only remember, aside from the one tap in, one shot (blocked) from #7 that was on frame. He is a wretched shooter.