Nick Collison, Undervalued?

[Levy2725]

I don't think he's undervalued by Thunder fans or the organization, but when it comes to the national media I think there is a lot of misunderstanding about the quality of his contributions to the team. I put together a little post at my site, Hickory-High, that I thought I'd share here. The link: http://hickory-high.com/2011/01/06/thunderheart.

I put together some statistics to back up my argument, here's a few of the main ones:

Since Collison returned from injury the Thunder have gone 18 - 9. If you're keeping track at home the Thunder's winning percentage went from 0.556 without him to 0.667 when Collison rejoined the team.

In November, after nine games, the Thunder had an ORtg. of 104.8 and a DRtg. of 107.7. In the 27 games since Collison returned the team has posted an ORtg. of 107.0 and a DRtg. of 103.4. This has brought their season ratings to 106.1 and 104.0 respectively.

In 548 minutes with Collison on the floor the Thunder have an ORtg. of 110.1 and a DRtg. of 102.2. In 1204 minutes without him their ORtg. is an identical 110.1 but their DRtg. rises precipitously to 110.0.

In terms of point differential they play like a 41 win team without him on the floor and a 61 win team when he is on the floor.

I'm primarily a Pacers' fan but the Thunder are one of the other NBA teams I watch consistently and there is something so satisfying about watching Nick Collison play. He's not flashy or jaw-dropping, but if you keep your eyes on him you'll find him consistently in the right place, making the smart play and doing the little things to help his team win.

I write for IndyCornrows, the Pacers' SBNation site, and there have been a lot of wild trade suggestions thrown out in the FanPosts looking for a defensive minded power forward to pair with Roy Hibbert. A few completely outlandish ones have centered around ways to get Ibaka and Harden to Indiana. I'm of the opinion that Collison would be a much better fit. I don't suppose we could interest you in a Solomon Jones/T.J. Ford pu-pu platter?

[slick_watts]

Nick has a lot of value to the Thunder because he's our most intelligent defensive player, and on offense he's probably one of the best screen setters in the league. Kevin Durant is at his best coming off screens and the difference between Jeff Green and Nick Collison setting those picks is huge. It's why most of our lineups featuring Durant and Collison perform so well offensively.

I am curious where you got your DRTG and ORTG numbers from. They don't look correct to me.

[slick_watts]

I figured out what's wrong with your numbers. You pull the team's ORTG and DRTG numbers from hoopdata gamelogs, which uses a simplified formula to calculate possessions that isn't stable. The relative rankings make sense but hoopdata's ORTG and DRTG figures are not compatible with the ones from 82games (where I assume you pulled the on/off data) or basketball-reference, both of those sites use Dean Oliver's method to calculate possessions which is more accurate.

i.e. in your article you state the Thunder's DRTG for the season to be 104.0 (which is pulled from hoopdata), but 82games.com has it as 107.0 using the more accurate possessions calculation.

[Levy2725]

I was just about to respond but it looks like you did all the legwork for me. I have noticed some differences in the past between hoopdata and 82games but didn't realize it was because of how they calculated possessions. Thanks for bringing that to my attention.

For this type of project I usually download lineup data from Basketball Value and manually calculate the ratings with players on and off the floor, but I was in a rush last night and just used the numbers from 82games.com.

Offhand do you know if the numbers on Basketball-Reference are more similar to 82games or Hoopdata?

[slick_watts]

Hoopdata is the only site I know of that uses the simplified formula to calculate possessions. basketball-reference and 82games both use Dean Oliver's formula.

I've run into this problem many times before when researching these types of things, I wish Hoopdata would make the switch because they are the only site with game logs that display ORTG and DRTG like that, which I know of.

[Levy2725]

Thanks again for the info. I'll really need to be careful with that.

Forgive me if this question reveals my mathematic ineptitude but since possessions are almost completely consistent with regards to offense vs. defense, the differences in the calculations would leave you with different ORtgs and DRtgs but the differential would be the same right?

[slick_watts]

Thanks again for the info. I'll really need to be careful with that.

Forgive me if this question reveals my mathematic ineptitude but since possessions are almost completely consistent with regards to offense vs. defense, the differences in the calculations would leave you with different ORtgs and DRtgs but the differential would be the same right?

Hoopdata calculates offensive possessions and defensive possessions separately for their offensive and defense ratings, which can lead to a different possessions figure for offense vs. defense. For example:

http://www.basketball-reference.com/box ... 60DAL.html

The formula for possessions that hoopdata uses is FTA-OREB+TO+FTA*.44

If you apply this for OKC's offense you get 88.44 possessions.
If you apply this for OKC's defense you get 89.80 possessions.

This will create small differences in differential over the course of a season compared to Dean Oliver's method which averages offense and defense possessions to give it more stability. i.e. using hoopdata's method OKC's differential for the year per 100 possessions is 2.1. Basketball-reference has it at 1.7.

[Devilanche]

I don't think he's undervalued by Thunder fans or the organization, but when it comes to the national media I think there is a lot of misunderstanding about the quality of his contributions to the team. I put together a little post at my site, Hickory-High, that I thought I'd share here. The link: http://hickory-high.com/2011/01/06/thunderheart.

I write for IndyCornrows, the Pacers' SBNation site, and there have been a lot of wild trade suggestions thrown out in the FanPosts looking for a defensive minded power forward to pair with Roy Hibbert. A few completely outlandish ones have centered around ways to get Ibaka and Harden to Indiana. I'm of the opinion that Collison would be a much better fit. I don't suppose we could interest you in a Solomon Jones/T.J. Ford pu-pu platter?

I think the offer is a non starter. Collison is in a weird scenario wage wise. At the moment, if you go over to the trade forum, there seems to be users who believe he's overpaid this season and that presti messed up by wasting our cap space this season on him. As such if he were to be traded this season it seems his value is as a salary dump for expiring but that doesnt take into effect his value from next season on. Next season he'll be very cheap salary wise and will probably only be packaged with a younger guy for a huge upgrade in our front court.

I don't think Pacer has anything that might interest Presti in a trade for collison.

[Levy2725]

My trade proposal was mostly tongue in cheek. I didn't really think the Thunder would be interested in trading him for a terrible team's two most ineffective players.

Watts - Thise may be another stupid question but does the possession difference you were telling me about before, that would effect all of the per possession stats that Hoopdata tracks, correct? In addition to throwing off their ORtgs and DRtgs things like TOR and AR would have the same issues?

[Devilanche]

i was kind of hoping you'll up your offer and throw in multiple picks or something.


What's happening to the pacers though? They started off great, Hibbert looks like he's paying off but granger seems to be having a down year, haven't heard much of darren too. And your posters on the trade forum seems to be in love with rush. Not even sure if paul george is starting though.

[Levy2725]

The Pacers are a mess right now. The offense has completely broken down. Collison is afraid to make a mistake and get benched so he's playing super hesitant. Hibbert looks exhausted, depressed and physically overwhelmed. Granger is trying to do to much and ends up hunting shots and killing the flow. Rush and Dunleavy have been playing well so George hasn't played much at all. The problem is Rush and Dunleavy aren't very effective if theother main options aren't hitting. Things are really falling apart.