Heej wrote:.
ShaqAttac wrote:.
OhayoKD wrote:.
Bad Gatorade wrote:.
Okay well since I’m being accused of cooking up fake numbers (despite fully providing sources and explaining the math), let’s start comparing numbers:
To cut to the chase:
AEnigma’s Numbers are Verifiably Contradictory with PBPstats’ Numbers (where I got my numbers from)
That’s not to say they’re wrong… but there’s a contradiction.
Part 1: Proof that Enigma’s Number’s Contradict PBPStats’ NumbersHere are the numbers that Enigma cites
-2017 Playoffs: +33.4 per 48 minutes
-2007-2008 Playoff Average: +25.8 per 48
-2016/2017 Playoff Average: +24.9 per 48
-2007-2010 Playoff average: +21 per 48
-2012-2017 Playoff average:: +14.7 per 48
-2015-2021 Playoff average:: +16.0 per 48
-2007-2012 Playoff average:: +15.4 per 48
-2014-2021 Playoff average:: +14.9 per 48
-2012-2021 Playoff average:: _+4.4 per 48
-2007-2021 Playoff average:: +13.8 per 48
Here are the numbers from PBP Stats ([url]https://www.pbpstats.com/wowy-combos/nba?TeamId=1610612739&Season=2016-17%SeasonType=Playoffs%PlayerIDs=2544[/url]):
-2017 Playoffs: 12.82 - -20.47 = 33.34 per 100 possessions
So right off the bat we have a contradiction for 2017. PBPstats’ NetRating follows the standard convention of using per 100 possessions, per the definition of net rating (
https://www.nba.com/stats/help/glossary).
You can use a player’s per 100 possession stats and their per 36 minute stats to adjust the units from possessions to minutes. LeBron’s Per 100 Possession -> Per 36 minute adjustment factor can be found by multiplying by the ratio of any of 2017 LeBron’s playoff stats per 36 mins with the same stat per 100 possessions. So for example: [10.5 FG per 36 mins] / [14.5 FG per 100 possessions] = 0.724 “per 36 minutes”/“per 100 possessions” (since the FG on top and bottom cancel out). You get approximately the same factor if you use any of the other LeBron stats (e.g. using TRB per 36 mins / TRB per 100 poss = 0.718) [source:
https://www.basketball-reference.com ].
Basic math: so 33.34 on/off “per 100 possessions” * 0.724 “per 36 mins” / “per 100 possessions” = 24.138 on/off per 36 minutes. You can adjust from per 36 minutes to per 48 minutes by multiplying by: 48 / 36. Since 48 mins is 1.333….x increase from 36 minutes, a stat per 48 is an 1.333….x increase from that stat per 36. For the record, this is the exact same procedure that Djoker cited for adjusting per 100 possessions to per 48 mins

So the final PBPstats’ on/off per 48 mins is:
-2017 Playoffs: 24.138 * (48/36) = 32.184 on/off per 48 minutes
This does not agree with AEnigma’s value of +33.4If we repeat the same process for Adjusting PBPstats for 2007 and 2008, we get:
-2007 Playoffs: 4.73 - -18.29 = +23.02 on/off per 100 possessions
-2007 Playoffs unit adjusted: 23.02 on/off per 100 possessions * (18.3 FGA per 36 mins / 25.0 FGA per 100 possessions) * (48/36) = 22.468 on/off per 48 mins.
-2008 Playoffs unit adjusted: (5.73 - -22.91)*(19.5./27.8)*(48/36) = 26.786
2007-2008 Playoff Average (averaging each year equally): 24.627 on/off per 48 mins
2007-2008 Playoff Average (averaging by game number): 24.169 on/off per 48 mins
2007-2008 Playoff Average (averaging by total minutes): 24.118 on/off per 48 mins
So once again, none of the PBPStats agree with AEnigma’s Numbers.If you continue this comparison, you’ll find many of the other numbers disagree.
Part 2: Possible Reasons for the ContradictionSo the first source of contradiction you might look if they get their values from a different source. I got my numbers from PBPstats and Basketball-reference, which are about as official and fully-vetted as you can get. The only mathematical change I did to the data was changing the units (which is fairly mathematical simple, and I’ve explained every step so anyone else can replicate the process). For multi-year stretches, I averaged by number of games as stated twice previously.
Meanwhile, AEnigma didn’t provide a source. To me more exact, AEnigma is claiming to have manually tracked LeBron’s on off by hand. Here’s what they said to me when I asked for the source for their numbers: “There is no “source”, you do the work yourself” by “replicating the process used for Jordan’s on/off (point differential and minutes).” Jordan’s numbers were tracked by hands from film. So AEnigma’s claiming to have tracked all the LeBron games by alone hand. AEnigma cited LeBron’s full playoff stats from 2007 all the way to 2021, which is a full 253 games
So this means:
AEnigma is claiming to have hand-tracked 253 LeBron games or over 200+ hours worth of footage, all on their own, without providing any proof of workTo be clear, these two methods produce blatant contradictions. They’re irreconcilable. There are only4 possible resolutions:
Possibility 1: PBPStats.com somehow accidentally got LeBron’s on/off numbers wrong, but just for LeBron (since the site is fully vetted otherwise).
Possibility 2: Basketball reference somehow accidentally got LeBron’s per 36 or per 100 number wrong, but just for LeBron (since the site is pretty fully vetted otherwise)
Possibility 3: Enigma made a mistake in their manual hand tracking of LeBron’s 253 playoff games… which we might be able to check, if Enigma hadn’t refused to provide any work when I asked them.
Possibility 4: Enigma just lied. They lied about manually tracking LeBron’s 200+ playoff hours by hand. They got the numbers from some website or made them up and lied about manually tracking. If it’s from a website, they haven’t as of yet provided that website so we’re unable to cross-reference that website with PBPStats to see where the disagreement lies.
…
Takeaway: I’m not to here to accuse another poster of outright trying. But Heej, if you’re going to accuse some poster of making up false numbers, you think it would be the one who claimed they manually tracked 253 LeBron games on their own and are refusing to show work, as opposed to the person who thoroughly provided sources for on/off per 100 possessions and explained the math for changing units in a way that anyone else could cross-check.
Now I’m not saying my numbers don’t have mistakes — if you find one, by all means point it out! But again, I’ve explained every source and step to recreate my numbers… while everyone else who’s claimed they’re false hasn’t explained why and hasn’t explained why we should believe their numbers more.
By the way OhayoKD, your numbers disagree with both mine and AEnigma’s. Can you please provide a source for yours so we can compare? On/off is defined by net rating / offensive rating / defensive rating, and it may be possible different websites have the slightest differences in those numbers… which may account for the difference