ACE: Need your help - suggestions for basketball UFR
As football season wraps up (I know, sadface), I'll be focusing a lot more on basketball, and Brian put forth the idea of coming up with a basketball version of UFR. I have a general idea of what I want it to include, which I'll list below:
- Rating shot quality: Contested jumper, wide open jumper, uncontested fast break layup/dunk, etc.
- Assigning points for good (and bad) plays off the ball, like picks that lead to layups, bad fouls away from the play, etc.
- Forced turnovers vs. unforced turnovers
- Defensive culpability—who, if anyone, was at fault for a given opponent basket
- Perhaps a rebound difficulty chart, resembling what Brian does with catches for Michigan receivers
I probably wouldn't post after every game, since that would be somewhat insane, and would also make it difficult for me to do anything else on the site, but I'm looking to put a hoops UFR through a test run during the Maui Invitational—since I'd still have FFFF to post on Wednesday, it might not go up until later in the week, but I want to start ironing out the kinks before Big Ten play begins.
Anyways, I would love to hear suggestions for what should go into these posts, and also any ideas for a name for this feature ('Basketball UFR' just feels like a copout) [never mind—UFR uber alles]. This would be focused on what doesn't show up in a box score or on KenPom, and yes, I will be going through play by play—this is where I thank the Big Ten for playing at a snail's pace.
So yeah, the floor is open for suggestions. I'd love to hear what you all think. Thanks in advance.
November 18th, 2011 at 12:22 PM ^
Man, that's going to be tough. Good Luck!
November 18th, 2011 at 12:29 PM ^
I don't know if it will be harder than a football UFR. For football, we chart every play, offense and defense. Many games, that's over 120 plays.
Basketball games are shorter, especially Big Ten basketball games. You won't have to chart every pass, just every possession.
November 18th, 2011 at 12:35 PM ^
there are so many more bball games though and not much time in between.
November 18th, 2011 at 12:38 PM ^
Oh, on a per-game basis, there's no comparison—I still have no idea how Brian does the football UFRs each week and actually manages to write, you know, other stuff. There's just far fewer things to look for—and fewer types of plays—in hoops. The only thing that makes a regular hoops UFR difficult is the sheer number of games.
November 18th, 2011 at 1:23 PM ^
Oh, good call, I didn't even think about that number of games part. Then yes, that will be a major pain for you.
November 18th, 2011 at 12:29 PM ^
Since blocks would show up in the stats.
November 18th, 2011 at 12:29 PM ^
For shots - knowing the type of defense (zone, man)
Title - Full Court Press
November 18th, 2011 at 12:35 PM ^
Full Court Press is the leader in the clubhouse right now. I like that a lot.
And yes, I should've mentioned that I'll have a similar format to football UFRs in terms of listing time, score, and type of defense, as well as noting the personnel on the floor. Any other suggestions for what should be charted for every play are very much welcome.
November 18th, 2011 at 12:30 PM ^
AABGWOCAR= Actual Analysis of a Basketball Game Without Crying About the Refs; Pronounced aab-gwoh-car.
A Maizin' Compilation
November 18th, 2011 at 12:33 PM ^
Call it The ACE (Assessing Culpability & Execution)
November 18th, 2011 at 12:40 PM ^
Well played. I'm not actually narcissistic enough to use this, but well played.
November 18th, 2011 at 12:34 PM ^
i think your first category - shot selection is most important. as a companion to shot selection, you should see what alternatives there were (i.e. the movement of the rest of the players, other players open for better shots, passing lanes, shot clock considerations, etc.)
good luck, this is a tough undertaking.
November 18th, 2011 at 1:58 PM ^
I think the shot selection is the most important too and maybe the most difficullt thing to track.
Especially hard for our team because we shoot lots of 3s. Sometimes I'm not sure if that was the best shot or if the team should have worked it around more or to the inside.
November 18th, 2011 at 12:37 PM ^
Ace, sorry this is off the topic. DO YOU HAVE A FACEBOOK? or atleast a picture somewhere.
November 18th, 2011 at 12:45 PM ^
I do have a Facebook, though I prefer to keep that private—I hope no one takes offense to this, but I don't accept friend requests from people I don't know (the internet doesn't count), mostly for privacy/security reasons. Facebook is pretty much the last internet outpost I have with some modicum of privacy.
You can find my picture on Twitter, though my avatar photo is actually around four years old. I've been trying to upload an avatar for here, but I still run into the old pair of error messages every time I try to do this. Help would be appreciated if you've found a workaround. Trust me, I'm not trying to hide or anything.
November 18th, 2011 at 1:04 PM ^
Totally understand. I was just curious because I do not want to referrence you just as "Ace", but hopefully one day we can meet in person. Go blue.
November 18th, 2011 at 2:29 PM ^
Not sure why you put it in quotes, considering it's his real name. Correct me if I'm wrong, but I'm pretty sure his actual name is Ace Anbender. Yeah his folks were cool like that.
November 18th, 2011 at 12:38 PM ^
"Basketball UFR" actually is a perfect name. Brian has said that UFR is the calling card of the blog, the central, main thing that it does around which most of the football analysis revolves. Therefore I think this is a thing where you should stick to the brand.
/marketing 101'd
I think with turnovers, you should look for the actual culprit instead of who the turnover is assigned to. For example, if Trey Burke throws a beautiful pass at Jordan Morgan in position for an easy bucket, but Morgan's not paying attention and it sails out of bounds, Burke will get the turnover but really it's Morgan's fault. Same if, say, a screener sets a poor screen and the dribbler goes where the defense isn't supposed to be, but gets stripped because a defender got around the bad screen. Turnover culpability could be a thing.
An off-ball chart would be huge. I mean, theoretically that's 80% or more of a player's time spent on the court. If a guy is a lazy screener, or great at defensive positioning when his man doesn't have the ball, that's important to know.
I would love to see a metric that gave us a player's true +/- impact. I think basketball +/- is a lousy stat that needs work, but it would take UFR-level drilldown to make it happen. A point scored is a point scored, IME, but a metric that gave a player's defensive value - points added for a block, points taken away for a crappy defensive play, that kind of thing - that would be a great thing.
November 18th, 2011 at 12:51 PM ^
Agree 100% re: branding - no need to get overly cute/quirky with what I would consider the meat & potatoes of this site. Leave that for the gravy.
November 18th, 2011 at 1:14 PM ^
I agree completely, it needs to be called UFR if it's going to be a basketball UFR.
November 18th, 2011 at 1:16 PM ^
bUFR! I'd read it!
November 18th, 2011 at 12:42 PM ^
Examining free throws made and missed as a quotient of a player's contribution seems like it would be useful. Driving to the lane and getting fouled is great, but if you can't make a free throw, it's tantamount to a turnover. And this team is obviously going to have trouble making free throws.
Also, it would be good to account for "hustle plays" like diving for loose balls, tipping out rebounds, taking charges, and general harassment, especially on the defensive end. Someone needs to keep track of all the charges Novak takes.
I feel like the most useful thing you can work on is devising a concise and vivid way to account for individual defensive contribution beyond blocks and steals. I know Stu is our best defender, in part because he's tasked with defending the opposing team's best player every night, but I don't really know why he's so good, what makes him good. If you can somehow put numbers on that, I would really appreciate it.
Ideas for UFR Titles:
Basketball 20/20 (as in "hindsight is ____")
Notes From The Block
The Intangibles Inquiry
The Back Screen
The Backdoor Cut
The Posterization of Zack Novak
Distributing the Rock
November 18th, 2011 at 3:15 PM ^
Fantastic suggestion. Analyzing only the possessions would take away judgement on where many games are won and lost -- the charity stripe.
November 18th, 2011 at 12:45 PM ^
Sounds like a great idea! Smotrycz would've gotten a minus one billion last night.
November 18th, 2011 at 1:17 PM ^
what's wrong with missing layups? jacking up 3s w/ 30 second on the shot clock?
November 18th, 2011 at 2:04 PM ^
To me he just looked a tad bit slower than he did last year. Maybe that is from the mass he cultivated in the offseason. I have faith in him to get into the flow and use his bigger frame to have a more complete game this season.
November 18th, 2011 at 12:47 PM ^
I don't know if we (meaning you) have any contacts on the staff that could share how they grade out games.
The thing that I like about the football UFR is that I feel our coaches could be using the same material. Or end up with similar results.
November 18th, 2011 at 12:47 PM ^
What about "A Beilein to the Truth" - you know, as it would presumably take one directly to what transpired during the game. Like a beeline. It's a pun. Not sure if there are issues with using a coach's name or not. Plus its easily abbreviated as BTTT.
November 18th, 2011 at 1:04 PM ^
If you can decipher my rambling:
Passes that lead to assists (aka hockey assists).
Assign bad pass turnovers to the player who threw the sloppy pass, not the fingertips of the intended recipient.
Pick/roll defense. Did players switch when they were supposed to? Did somebody not fight through the screen? Was the pick not called out?
Grade a player negatively for giving up and-1s when a harder foul would have been necessary.
If a player is double teamed does he look for the open man or does he force a play?
Keep an eye out for guys cheating in the passing lanes and getting burned.
Credit charges! Also, if you could, point out any unnecessary flops that compromised the defense.
Failure to box out leading to missed rebounds and/or cheap offensive putbacks. Conversely, successful boxouts that paved the way for somebody to snatch up an easy rebound or drew an over the back would be nice.
I'll think of more.
November 18th, 2011 at 12:48 PM ^
Is there anyway you could just calculate each player's player efficiency rating (PER) as John Hollinger does for the NBA. I believe these rankings could give a great estimation to quantify a player's worth each game and also look a how that number changes over the course of a season. I believe that this could save a lot of time as I am sure all of the data can just be plugged into some Excel spreadsheet after each game to generate the PER value.
November 18th, 2011 at 12:56 PM ^
I love advanced basketball stats, but KenPom does such a good job with them, and the point of these posts would be to try to quantify the stuff that doesn't show up in any box score—much like how Brian's UFR is supposed to complement, not replace, advanced statistical measures and also your own impressions of the game.
Also, while season-long PER ratings are very significant, doing a rating for an individual game will have a lot of noise and probably wouldn't tell us too much beyond the obvious 'did he play well?' anyway.
November 18th, 2011 at 12:48 PM ^
This would be incredible if you could pull that off, and IME would allow MGoBlog to overtake UMHoops for best comprehensive Michigan basketball coverage. If you only did it for one game a week come Big Ten season, that might be a more reasonable approach than attempting every single game (IME that would require having a 2nd person do all of one side of the ball).
Rate shot quality, along with docking points if there was a better shot or passthe player could have taken or made. Consider the player taking the shot; an extreme case is if a poor 3-point shooter is the one taking the shot, and using poor form to do so. Consider also the circumstances, like dealing with fast-break opportunities vs half-court sets.
I look forward to seeing what you come up with for the Maui Invitational, and how it could evolve over time.
November 18th, 2011 at 1:17 PM ^
I'd think it feasible to do an analysis of highlight standard packages so he wouldn't have to watch every play of every game but could still provide insight as to why such and such a high/lowlight came to be.
November 18th, 2011 at 1:42 PM ^
You're talking about basketball picture pages right?
November 18th, 2011 at 12:50 PM ^
I think it'll be important to track shots created. For example, give Trey Burke a point for a well-executed drive and kick that results in an open jump shot, even if that shot isn't made. Likewise, account for open shots that players create for themselves (e.g. Tim Hardaway creating an open step-back jumper for himself) vs. shots that fall into players' laps (from something like the above-mentioned drive and kick).
November 18th, 2011 at 12:54 PM ^
It seems like we are going to play a lot of different guys this year and that Beilein is still experimenting with putting guys together in different combinations and putting guys like Novak at different positions. I think some sort of detailed +/- analysis by lineup grouping might be interesting, especially with this team. Do we defend better or accomplish more on the glass if we go big? Do we score more if we go small and does that make up for any losses defensively? Does it help to have Burke and Stu on the floor at the same time to give us more ball handlers?
Might be neat and might contain some of the other stuff you would have focused on and allow you to save a little time (seriously, this sounds like it could ruin a person's life if you did it in too much detail).
November 18th, 2011 at 1:15 PM ^
On defense, I think it would be good to keep track of shots altered. You could tally especially effective close-outs on the perimeter, as well as interior shots that are missed out of intimidation (this will probably be more relevant next year; looking at you, Mitch McGary).
For the record, I think a basketball UFR is a terrific tremendous idea.
November 18th, 2011 at 12:55 PM ^
is understating the undertaking here. While there are theoretically "plays" and Coach B obviously has a scheme, a lot of basketball is instant on court decision making and isn't amenable to a UFR-type analysis.
This sounds like it may be more of a sabremetrics or stats-that-really-matter type approach, which will be awesome in its own right. Outside the Box, perhaps?
November 18th, 2011 at 1:02 PM ^
Kidding. . . but I'm a bit of a basketball nerd and first you'd have to think of all the thing you want to quantify. I'll just throw somethings out there on offense and defense.
Offense:
Shot making. Probably difficulty of shot and how appropriate the shot was. two different metrics. sometimes because of the shot clock or situaion a difficult shot is needed although usually it's not called for. Shot creation might also be a third metric here to see which players are able to iso their defender.
Moving without the ball. Setting picks, and getting open (whether or not they get you the ball) would be easily quantifiable
Creation. A great pass that leads to a bucket should count the same as one that leads to a missed bunny or a foul shots. A bad pass, or missed open player could potentially count against the passer.
Hustle/Derp metric. Here we have level of offensive rebounding, saving or diving for a ball, and offensive fouls or turnovers. Turnovers could also go under Creation.
Defense:
Rebounding: Who contributed to the rebound whether it was boxing out or out jumping someone.
Shot making of oppent: Who's responsible for high quality shots or someone breaking down the defense?
Assist defense. Who's responsible for easy shots that were created by the other team or potential assits that were defended?
Pick n Roll defense: The hardest play to defend so maybe it needs its own metric.
Fouls: There are good fouls and bad fouls and bad calls either way. A metric can be created to quantify this.
Overall an offensive rating and a defensive rating can be calculated for each player and if you want to parse out hustle/rebounding then you would have potentially a third rating.
November 18th, 2011 at 1:09 PM ^
I think if you do it, the obvious thing is to chart each possession the way Brian charts plays. But one major change would be that I think you'd have to list offensive and defensive possessions consecutively. The flow of the game is such that doing an offense UFR and a defense UFR would be confusing. I'd suggest highlighting Michigan possessions in blue (a la current UFR), and opponent possessions in whatever their main color is.
November 18th, 2011 at 1:49 PM ^
Yeah, I think the unit of analysis has to be a "possession", and possession resets with every change in possession (duh) and every shot attempt.
Lots of good suggestions already here. Like Brian scores Denard on the zone reads, you'll need to have a metric of "shoot or pass" (or shoot or pass or dribble).
I look forward to your Maui trial run -- after you do it you'll have a sense of what you like/don't like to follow, and then you'll get reaction-posts as opposed to "anything in the ether" posts.
November 18th, 2011 at 1:10 PM ^
That should be the title. Just sayin.
November 18th, 2011 at 1:14 PM ^
A good pass that leads to free throws and a pass that leads to a pass for an assist, IE a skip pass. Another thing that might be hard to track but helps you win games is getting a 50-50 ball.
November 18th, 2011 at 1:15 PM ^
Make it an MgoBlog institution.
November 18th, 2011 at 1:20 PM ^
Upon further review (i.e., a quick chat with Brian), we're going with UFR. So yeah, MGoBlog institution it is.
November 18th, 2011 at 1:29 PM ^
Call it the Buffer
November 18th, 2011 at 1:21 PM ^
But I like a few ideas from above:
Hockey Assist
Altered Shots
Good/Bad Fouls
Adjusted Shooting Percentage ideas (Quanitfying shot selection. Do all the end-of-shotclock-desperation-throws need to count in shooting percentage? Same idea that Denard's End of half hail-mary is an INT... but not an important "miss".)
November 18th, 2011 at 1:23 PM ^
My advice.
I'm a former basketball player, so I know more about schematics here than I do about football. I'm not sure what the end result of what you want to do is here. The numbers that traditionally tell you who is being effective (points, rebounds, assists, steals, charges, FT %) are pretty straightforward.
Further, unless you specifically know the play calls, how are you going to judge who is doing what correctly? On defense, if you aren't sitting in the pre-game planning session and don't know how, for example, Beilein plans to defend Sullinger (perhaps switching possession to possession), how can you judge it beyond the cosmetic and the known (FG %, blocks, etc.)
I'm just not sure you want to go down the road of deeming who is playing well and who is not based upon what you think you're seeing, as opposed to what the coaches and players actually know they're to be doing. Despite what it might look like, basketball is an exceedingly complex game of technique and precision.
November 18th, 2011 at 1:30 PM ^
This is a point of view I very much understand, but unfortunately, it's not like I can be in the pre-game meetings—you could point to the exact same issue for football UFR. I know basketball is complicated, and I don't know everything that Beilein wants his guys to do out there, but there are still things that can be pointed out based on a general knowledge of the sport.
These posts will be tweaked and altered over time, much like the football UFRs, as I learn more about which metrics work best and which ones either don't provide good information or are simply superflous. I'm always open to suggestions and input, especially from players/coaches with a deeper understanding of the game. I just don't think it makes sense to scrap the idea entirely because I'm not a player or coach—as long as my limitations are acknowledged, and people understand that this is a tool for analyzing the game, and not the be-all-end-all of game breakdowns, then I think it can be very valuable.
November 18th, 2011 at 1:36 PM ^
Wouldn't basketball be the same as football in the way you can infer on what's suppose to happen. If you have a high football IQ you understand what is suppose to happen on a given play while watching the game. Now I don't have a great basketball IQ but why can't you do the same? I can look at basketball games and understand who messed up where and who was out of position and so on, so I would imagine people with a much better IQ can deliver a product like Ace is saying. Or am I totally off base here?