Winshares,winshares/48 and on/off metrics

[migya]

Of all the advanced metrics winshares seems to be the best at indicating the value of a player. Contribution to winning is the aim of any player and the team. An owner of a franchise wants to win the most possible as that is success in basketball.

On/off stats are often mentioned as a good indication of the value of a player. It seems logical that the higher the number in this metric the higher the value of a player and thus the higher the contribution to winning.

That is not the case with some players. This is the case with Garnett who has among the highest on/off numbers and relatively low winshares and winshares per 48 minutes, particularly compared to other great players.

What is the view in relation to these statistics?

[70sFan]

Have you checked how win shares are calculated?

[migya]

Have you checked how win shares are calculated?

Yes, again you veer from the topic.

[ceoofkobefans]

Funnily enough WS (and especially adjusting it to /48 minutes) is up there for one of the worst impact metrics there is lol. I love WS so I try not to **** on it but it is based on the player versions of ORTG and DRTG which makes it one of the worst metrics at evaluating defense and it’s ok at evaluating offense. Some of the best impact metrics are EPM DARKO DPM RAPTOR And everyone’s favorite RAPM (although needs a large sample). Historically RAPTOR and Back Picks’ version of BPM are probably the best impact metrics you’ll get

[70sFan]

Have you checked how win shares are calculated?

Yes, again you veer from the topic.

You created a topic about win shares, how is my question not about the topic?

Win shares try to calculate the contribution of individual player into team success. The problem is that it does quite poorly in actual tests.

[migya]

Have you checked how win shares are calculated?

Yes, again you veer from the topic.

You created a topic about win shares, how is my question not about the topic?

Win shares try to calculate the contribution of individual player into team success. The problem is that it does quite poorly in actual tests.

How?

[70sFan]

Yes, again you veer from the topic.

You created a topic about win shares, how is my question not about the topic?

Win shares try to calculate the contribution of individual player into team success. The problem is that it does quite poorly in actual tests.

How?

You can read about it here:

https://fansided.com/2019/01/08/nylon-calculus-best-advanced-stat/

[LukaTheGOAT]

WS/48 doesn't do great in testing as already mentioned. Here is one example

Read on Twitter

https://dunksandthrees.com/blog/metric-comparison?s=09

NBA Executives when asked about the best all-in-one metric:


What is the best advanced statistic for basketball? NBA executives weigh in https://hoopshype.com/lists/advanced-stats-nba-real-plus-minus-rapm-win-shares-analytics/

Also this is off-topic, but not quite sure about the "An owner of a franchise wants to win the most possible as that is success in basketball,"....Just look at Jeanie Buss and the Lakers lol

[eminence]

WS are pretty solid for what they are, and I'm quite happy to have such an all in one for earlier eras, these days we usually have better options.

[MiamiBulls]

Win Shares is fairly junky and archaic. One thing that's very obvious about Win Shares is that WS overweights Rebounding specifically Defensive Rebounding.



See: WS & WS/48 of '23 Nets Kyrie Irving compared to Nic Claxton. WS & WS/48 of '23 Cavs Darius Garland compared to Jarrett Allen. WS & WS/48 of '23 Celtics Jaylen Brown compared to Al Horford. WS & WS/48 of '23 Clippers Paul George compared to Ivica Zubac. The pattern is very clear.

WS, WS/48 is about as low grade in terms of composite data as PER. It had value in 2014, but 2023 isn't very useful in terms of gauging value of its teams top players.

[eminence]

Win Shares is fairly junky and archaic. One thing that's very obvious about Win Shares is that WS overweights Rebounding specifically Defensive Rebounding.



See: WS & WS/48 of '23 Nets Kyrie Irving compared to Nic Claxton. WS & WS/48 of '23 Cavs Darius Garland compared to Jarrett Allen. WS & WS/48 of '23 Celtics Jaylen Brown compared to Al Horford. WS & WS/48 of '23 Clippers Paul George compared to Ivica Zubac. The pattern is very clear.

WS, WS/48 is about as low grade in terms of composite data as PER. It had value in 2014, but 2023 isn't very useful in terms of gauging value of its teams top players.

I agree with the general point that WS heavily weights rebounding, but in your particular examples I can see decent arguments either way in 3/4 teammate pairs for this season (not Zubac/George).

[Dutchball97]

Win Shares is fairly junky and archaic. One thing that's very obvious about Win Shares is that WS overweights Rebounding specifically Defensive Rebounding.



See: WS & WS/48 of '23 Nets Kyrie Irving compared to Nic Claxton. WS & WS/48 of '23 Cavs Darius Garland compared to Jarrett Allen. WS & WS/48 of '23 Celtics Jaylen Brown compared to Al Horford. WS & WS/48 of '23 Clippers Paul George compared to Ivica Zubac. The pattern is very clear.

WS, WS/48 is about as low grade in terms of composite data as PER. It had value in 2014, but 2023 isn't very useful in terms of gauging value of its teams top players.

I agree with the general point that WS heavily weights rebounding, but in your particular examples I can see decent arguments either way in 3/4 teammate pairs for this season (not Zubac/George).

EPM has it like this:

Nic Claxton +3.7 > Kyrie Irving +2.9
Jarrett Allen +2.9 > Darius Garland +2.4
Jaylen Brown +3.5 > Al Horford +1.5
Paul George +5 > Ivica Zubac +0.5

Further comparing them through RAPTOR and BPM we see Claxton having 0.1-0.2 leads in both stats suggesting they're close but that Claxton is likely the slightly better player this season. Garland has a pretty significant edge in RAPTOR and a similar advantage in BPM as Allen has in EPM so this one could go either way but leaning Garland. Horford has a decent lead in RAPTOR and a big lead in BPM with Brown of course having the large EPM advantage, this one I'm honestly not sure about. George of course also far outperforms Zubac in both RAPTOR and BPM.

[migya]

Win Shares is fairly junky and archaic. One thing that's very obvious about Win Shares is that WS overweights Rebounding specifically Defensive Rebounding.

Yet Garnett has low ws/48 compared to most star bigs and he rebounds among the best.

[migya]

Winshares is consistent among the stars. Chamberlain, Kareem, Robinson, Shaq, Malone, Barkley are some examples. I don't understand why, for example, it's considerably lower for Garnett while his on/off is high.

[Dr Positivity]

KG style of defense doesn't show up as much in steals and blocks which slightly deflates his WS, for example in 08 he is only 0.5 DWS ahead of Pierce. But I think they're still fine, from 03-06 in WS/48 he finishes 5th, 1st, 2nd, 3rd, and then 2nd in 08. Nash is more underwhelming at 15th, 10th, 5th and 14th from 05-08 which is more like what he was doing with the Mavs. It seems like his per possession stats are clearly better with the Suns so my only explanation would be that the rest of the league was putting up good offensive stats as well in post handcheck era. In 2003 Jamison is 9th in OWS putting up a 22ppg .542 TS% season, Hawks Shareef is 12th with 21ppg .536. In modern day Tatum's 31 pts on .613 TS% is only 24th in OWS. Mavs Nash is probably a little bit underrated in a more difficult era to put up stats.

[rk2023]

Of all the advanced metrics winshares seems to be the best at indicating the value of a player. Contribution to winning is the aim of any player and the team. An owner of a franchise wants to win the most possible as that is success in basketball.

On/off stats are often mentioned as a good indication of the value of a player. It seems logical that the higher the number in this metric the higher the value of a player and thus the higher the contribution to winning.

That is not the case with some players. This is the case with Garnett who has among the highest on/off numbers and relatively low winshares and winshares per 48 minutes, particularly compared to other great players.

What is the view in relation to these statistics?

Hello Migya,

It seems you are a sage mind when it comes to leveraging basketball reference as a source of analysis. Being one who hasn't toggled with the platform as much, am curious as to how Chris Paul stacks up to Stephen Curry and Magic Johnson with that as a point of reference - and how would you rank (arguably) the 3 best points guards of the last 40 years? Your intel would be appreciated

[OhayoKD]

You created a topic about win shares, how is my question not about the topic?

Win shares try to calculate the contribution of individual player into team success. The problem is that it does quite poorly in actual tests.

How?

You can read about it here:

https://fansided.com/2019/01/08/nylon-calculus-best-advanced-stat/

You may have seen this before, but for posterity, this study builds off ben's base(or at least claims to):
https://dunksandthrees.com/blog/metric-comparison

Funnily enough WS (and especially adjusting it to /48 minutes) is up there for one of the worst impact metrics there is lol. I love WS so I try not to **** on it but it is based on the player versions of ORTG and DRTG which makes it one of the worst metrics at evaluating defense and it’s ok at evaluating offense. Some of the best impact metrics are EPM DARKO DPM RAPTOR And everyone’s favorite RAPM (although needs a large sample). Historically RAPTOR and Back Picks’ version of BPM are probably the best impact metrics you’ll get

Eh........RAPTOR and BPM don't have historical plus-minus data pre-97 which means they're basically just ws/48 or PER when it comes to "historic" players.

You're probably better off using WOWY(and then applying knowledge to account for fit, team context, ect). At least for players we don't have other methods for(jordan has on/off and some rapm from his peak, Hakeem has some RAPM in off-years).

[Owly]

Winshares is consistent among the stars. Chamberlain, Kareem, Robinson, Shaq, Malone, Barkley are some examples. I don't understand why, for example, it's considerably lower for Garnett while his on/off is high.

Because it does not use play by play data. It doesn't know that the T-Wolves were (97-07) +3.7 with him on and -8.5 (or that for a couple of years they were 20 points worse per 100 with him off the floor), merely that that the T-Wolves weren't that good. It's unlikely for a player that good to be on a team that bad, but it happens and WS hits Minny era KG for it.


I'm inclined to suggest, to my understanding at least that WS/48 (and WS in general) is not notably in love with rebounders (a Nater or Reggie Evans's numbers don't look crazy versus, say PER [BPM's interplay of factors makes it a touch more complex]) moreso it will like the efficiency of a Claxton (123 TS plus) especially and basket-advacent/dunker/rim-runner big with little regard for shot creation.

[ty 4191]

Which players have the highest career on/off in the playoffs, all time?

[OhayoKD]

Which players have the highest career on/off in the playoffs, all time?

Going off ben's videos(note it doesn't include non jordan pre-97 players), going by 3 year stretches you have Duncan/Drob topping, 16-18 Lebron second(among the players he shows), and then curry, shaq and jordan tied for 3rd iirc(shaq actually shows up at 2nd but then unibro and 70's found ben made a statistical error.