APBRmetrics

First off - thanks to some of the guys from here that now follow me on twitter - I'm trying to get in the swing of this whole modern media/blog thing, it's all very new to me. I want to follow any of you that debate/reveal any basketball analytics - hit me up here with your twitter name or just follow me on twitter if you'd like, and I'll reciprocate.

Next - I've been on a hiatus again for a while from my ratings (life, what canya do) - BUT I did just recently finish my version of NBA WAR. Please check it out - let me know what some of you may think about my methodology (like maybe most controversial, no season player WAR is a negative) or the results. I'll be doing the 2013 playoffs WAR soon (HOPEFULLY Sunday/Monday) - and I now am excited about doing the historical seasons, hopefully I get a few days off work and marathon it.

Finally, I am fully debating actually updating my NBA player WAR every single day of this upcoming season. I'll post updates on twitter - I'll probably include change in WAR (and rank that result also) over the last day, week, and month. This will be very tough considering my hectic schedule, but I work better under pressure. Right now it's a go - unless a life (career) change happens that nixes that plan.

...maybe most controversial, no season player WAR is a negative...

I don't see this as very controversial. If a player is worse than replacement level, he'd be replaced.
A few players in eWins are negative, but they don't get many minutes. I figure they're "projects" who fail to develop.
I'll be looking at your stuff, and I expect you to check into the Voting threads here, in your extensive free time.
Guessing that you don't have LeBron with any 30 wins, as I've seen others crediting him with.

EDIT - haha, I guessed wrong.

Mike G wrote:

...maybe most controversial, no season player WAR is a negative...

I don't see this as very controversial. If a player is worse than replacement level, he'd be replaced.
A few players in eWins are negative, but they don't get many minutes. I figure they're "projects" who fail to develop.
I'll be looking at your stuff, and I expect you to check into the Voting threads here, in your extensive free time.
Guessing that you don't have LeBron with any 30 wins, as I've seen others crediting him with.

EDIT - haha, I guessed wrong.

28.52 isn't quite 30.......

BTW, should someone expect last year's Heat to win more than 38 games with LeBron's minutes replaced with replacement level minutes? Parts of me almost thinks 28.5 WAR might be a smidge low.

BTW, should someone expect last year's Heat to win more than 38 games with LeBron's minutes replaced with replacement level minutes?

yes, absolutely... because with james gone the touches/min on offense of both wade and bosh assuredly go up...

the 3 years with james (10-11 to 12-13) both wade and bosh saw 20% drops in their touches/min on offense from their 3 previous seasons (07-08 to 09-10)...

Furthermore, if LeBron goes down, we wouldn't see 100% of his minutes given to guys off the street. There are a number of better-than-replacement-level guys who would actually replace most or all of his minutes.

I doubt that there's ever been a team that is one player away from needing to go with 3000 replacement level minutes.

Mike G wrote:Furthermore, if LeBron goes down, we wouldn't see 100% of his minutes given to guys off the street. There are a number of better-than-replacement-level guys who would actually replace most or all of his minutes.

I doubt that there's ever been a team that is one player away from needing to go with 3000 replacement level minutes.

Agreed - but WAR is wins above "replacement". If I did WAAJBAP (wins above a just below average player), then obviously the result would be a bit lower than 28.5.

As for increased touches for Wade/Bosh - yes, true. You'd also have increased touches for not as efficient players also - as well as a lineup that loses its most efficient scorer, best passer (hurting efficiency of teammates), its best defender (against any position), and one of its better rebounders.

NO James at all, oldish Wade & Bosh (missing some games themselves) with 3000 minutes played by a replacement player, and the rest of the minutes played by guys that - for the most part - are now below average NBA guys - I personally still don't see a team close to .500.

Mike G wrote:

...maybe most controversial, no season player WAR is a negative...

I don't see this as very controversial. If a player is worse than replacement level, he'd be replaced.
A few players in eWins are negative, but they don't get many minutes. I figure they're "projects" who fail to develop.
I'll be looking at your stuff, and I expect you to check into the Voting threads here, in your extensive free time.
Guessing that you don't have LeBron with any 30 wins, as I've seen others crediting him with.

EDIT - haha, I guessed wrong.

Tell that to Austin Rivers.

If a team had only replacement level guys, they'd win on average X wins (what is it these days? 10 wins?) But obviously you can play worse than replacement level. So I don't see why you don't want below replacement value.) Though I guess it depends on what you define as replacement level and how many wins that team would have with only those guys.

Mike G wrote:Furthermore, if LeBron goes down, we wouldn't see 100% of his minutes given to guys off the street. There are a number of better-than-replacement-level guys who would actually replace most or all of his minutes.

I doubt that there's ever been a team that is one player away from needing to go with 3000 replacement level minutes.

The LeBron Cavs? That's probably one of the closest. They were usually shallow in wing depth. They willingly played Pavlovic, and 2008 Szczerbiak was injured a lot (I think.) That's definitely one of the biggest differences between top layer with 3000 minutes and replacement options.

AcrossTheCourt wrote:Tell that to Austin Rivers.

If a team had only replacement level guys, they'd win on average X wins (what is it these days? 10 wins?) But obviously you can play worse than replacement level. So I don't see why you don't want below replacement value.) Though I guess it depends on what you define as replacement level and how many wins that team would have with only those guys.

I actually deliberately set replacement level a little low (80, when league average is 100) - at a level that is truly available for 10 day contracts. Yes, there are players that played VERY poorly - I actually predicted poor Austin Rivers to be the worst value pick of the entire draft in a write up I did last season - but those below 80 HnI are just plain replacement level and their WAR is zero.

It makes no sense to me to have poor Austin at some fairly solid negative WAR - if he truly was much worse than replacement, his coaches (and the guys that decided to draft him) would have to be some serious morons giving him all those minutes. Plus - getting REAL minutes in the NBA (at whatever level your metrics say) shouldn't make your career WORSE than some guy that played 1 minute.

Statman wrote:It makes no sense to me to have poor Austin at some fairly solid negative WAR - if he truly was much worse than replacement, his coaches (and the guys that decided to draft him) would have to be some serious morons giving him all those minutes.

Speaking as someone who has Rivers rated so poorly, I would disagree. First, if the experience he's getting potentially translates into future above-replacement play, that's a reason to continue playing him despite the short-term ramification. Also, to say that no player can play below replacement is to ignore a lag time in evaluation and sample-size issues. Just because a player rates below replacement level during a given timeframe doesn't mean he's truly sub-replacement talent.

NO James at all, oldish Wade & Bosh (missing some games themselves) with 3000 minutes played by a replacement player, and the rest of the minutes played by guys that - for the most part - are now below average NBA guys - I personally still don't see a team close to .500.

not even close to .500? are you seriously trying to suggest that a team that went 66-16 with lebron james (27 pts/g, 2.71 pts/0ptposs, their best defender) would go less than .500 without him?...

from 90-91 to 92-93 the chicago bulls averaged a 62-20 record (michael jordan 31 pts/g, 2.72 pts/0ptposs, their best defender). from 95-96 to 97-98 they averaged a 68-14 record (michael jordan 30 pts/g, 2.51 pts/0ptposs, their best defender)...

but in 93-94 without jordan (but with pete myers) they went 55-27, and in 94-95 went 47-35 (jordan just 17 games, shot 40% on 2s), well above .500 both seasons...

how many games does your methodology say the 93-94 and 94-95 bulls should have won without jordan?...

BobC and I have both failed to note that removing a player's WAR refers to 2 hypotheticals:
1 - that the player is absent
2 - that the player's minutes are taken by truly replacement level players

In almost every case, when a starter goes down, other starters play more and do more when they play. The 6th and 7th men play a lot more -- and sometimes these guys are starter quality -- and so on; perhaps giving more than mop up minutes to the actual projects and practice guys on the bench.

The Bulls in '93 thru '95 of course didn't give all of Jordan's minutes to Pete Myers; while he was a 'replacement' in the starting lineup, he wasn't the type that any team could find several times over.

The Bulls also happened to pick up Toni Kukoc, Steve Kerr, Bill Wennington, and Luc Longley in '94; Ron Harper in '95.

I don't really care for this Replacement Level business, but I'm responding to the point made about a LeBron-less Heat.

If healthy, a Heat team without LeBron would likely be a 3-seed in the East. People forget that when running an offense, Dwyane Wade is nearly as effective a play-maker as James. He would likely put up 25 PPG to boot, albeit on a somewhat lower TS% (55% compared to his current 57%). Bosh would likely raise his scoring output to 23 PPG, and based on the on/off numbers I've seen, he could likely due it without a hint of lower efficiency. Not to mention players like Allen, Birdman, and Oden (if healthy), who could all stand to take a few more shots.

The only significant drop-off you would see would be in the Playoffs, where I could see the Heat bowing out in the 2nd round. But to say they would be sub .500 is kind of crazy.

bchaikin wrote:
how many games does your methodology say the 93-94 and 94-95 bulls should have won without jordan?...

I'll tell you when I do it - it'll be a little while before I get the historical stuff done back to the mid 90s.

I can also play the "how did a team do w/o their superstar player the following year" game btw.

The Cavs w/ LeBron: 66 wins when he played 81 games, 61 wins when he played 76 games, then 19 wins when he played for Miami.

HoopDon wrote: The only significant drop-off you would see would be in the Playoffs, where I could see the Heat bowing out in the 2nd round. But to say they would be sub .500 is kind of crazy.

Once again - I stated it would be hard for me to imagine LAST season's Heat w/ 3000 some odd LeBron minutes replaced by "replacement level player" getting to .500.

To say they would be a 3 seed - and the only "significant drop off" would be in the playoffs w/o LeBron seems just as crazy to me from the other end of the spectrum.

These aren't the '09 versions of Wade, Bosh, & Allen. It seems assumptions are being made that somehow these guys would magically be the 4 year younger non injured versions of themselves w/o James hogging all the possessions.

To think the Heat WITHOUT James would still be a bit better than, say, the Celtics last season - that seems a stretch to me. To think w/o James they'd be as good as the Bulls without Rose seems to insinuate that James really isn't really any better than Rose. That's quite a stretch for me also.

Maybe that's just me. James is such an outlier - his WAR is so big - I'm not surprised it triggers such debate. The number doesn't seem that outlandish to me - but I also look at LeBron as being probably as good or better than any player in history. Obviously others don't feel as strongly about his impact as I.

This just makes me more interested to see how the historical WARs pan out.

There are some interesting patterns I see:

1. You are operating under the assumption that a replacement level team would win 0 NBA games in a season... In baseball, a replacement level team has an estimated winning percentage of something like 29%.... There have been teams in MLB history that have win at that rate or worse. Obviously 29% is too high for basketball, but if you took the 10 worst records in the history, one could choose maybe something like 12-15% winning percentage (10-12 wins or so)

2. Your average WAR is about 2.6 wins (assuming I have no typos), and median is 1.13 wins.... In baseball they are close to zero... Now, I know in baseball there are more players on a team, so maybe that's the reason.


Either way, I think if the assumption was a replacement level team would win some games, I think the WAR numbers for LeBron for example might come down (even though it would still be the highest).

Looking at basketball-reference, LeBron's win shares are about 15-20 depending on the season. That just feels more right, than saying 28 or 30, IMO.

HoopDon wrote:I don't really care for this Replacement Level business, but I'm responding to the point made about a LeBron-less Heat.

I'm curious - do you feel the same about WAR in baseball?

Ignoring what the WAR "numbers" say (ie, LeBron at 28.5) - do the rankings of of the players by WAR make sense to you - or do they seem out of wack? Are there players who you think should be ranked much higher or lower in general.

I'm just wondering - I'm not trying to be any way flippant. The point of my NBA WAR isn't to be the exact predictor of how a team would have done w/o a player - it was a way for me to have a rating that ties directly to player performance and team success - and compiles nicely when combining seasons for comparisons. The big differences in last season's WAR between #1, #2, and #3 just tells me that there was an obvious best player in the NBA, and obvious 2nd best, and a "who really knows for sure" 3rd best (Paul would be an obvious 3rd best if he played more). My subjective experience and perceptions of players last season fits fairly well with the WAR rankings - which is why I am more than comfortable with publishing the results, compiling historical results soon, and producing ongoing WAR this next season day by day.