Offensive PER's of the Rockets

Of coz it includes FT attempts, or more accurately FT made, unless the scores from FT are not included in the total score ...

 

As far as easier goes, you will need a formula that weights games played, too. Not just mpg

See how complicated this gets like trying to make +/- individual stats equal true team pt differential

 

Weighted for total minutes played not mpg. That would be foolish. Is that not what the OP did?

EDIT: Oh it does appear that's what he did. He used MPG not total minutes played. That's what I assumed because that's the only way this makes sense. I sort of understand your hang-up now, heyp, because you thought it was mpg. Very simple solution for needing "a formula that weights games played", just used total minutes played .

 

See if you can convert player +/- into true team differential first. I think what you are trying is more complicated than needs to be. Just take the team boxscore

 

You're really not getting this math are ya? Let's make it simple and say PER only consists of PTS+REB+AST and call it sPER (simple PER). Let's say player A got 10 points, 10 rebs, 10 assists in 30 minutes. His sPER would be (10+10+10)/(30 minutes) = 1 sPER. Let's say player B had 40 points, 40 rebs, and 40 assists in 60 minutes. His sPER would be (40+40+40)/(60 minutes) = 2 sPER.

Your way of calculating their average PER using the entire box score would be:

(10+10+10+40+40+40)/(30 minutes + 60 minutes) = 1.67 sPER

My way of just using weighted minutes would just be:

((1 sPER*30 minutes) + (2 sPER*60minutes))/(30 minutes + 60 minutes) = 1.67 PER.

Of course PER is way way more complicated and the numerator part of the first calculation would be way way more complicated whereas the second one you can just take the published PER and just mess with the minutes. Which one is simpler?

I really hope this helps you understand my point.

 

PER is one of the most flawed offensive metrics there is. Its biggest flaw is rewarding volume shooting. Take example of Rudy Gay, 38% FG. His PER is 17.04, an above average offensive player according to PER. A team full of Gays should do well, right? Not so!

If the whole team has the same efficiency as Gay, they are GUARANTEED to be the worst team in the league!

PER can also be manipulated, especially on a contract year. Players like Blatche and Hickson aren't great players by any standards, but for years, they looked like stars by PER metrics. Some players are just stats whores and PER will not able to discriminate the difference.
<iframe width="420" height="315" src="//www.youtube.com/embed/lDzRvZcFn48" frameborder="0" allowfullscreen></iframe>

 

Here is the correct team PER for the Rockets so far:

<table class="tableizer-table"><tr class="tableizer-firstrow"><th>Rk</th><th>Player</th><th>MP</th><th>PER</th><th>PER*MP</th</tr>
<tr><td>1</td><td>James Harden</td><td>398</td><td>21.6</td><td>7681.4</td></tr><tr><td>2</td><td>Jeremy Lin</td><td>372</td><td>19.3</td><td>5914.8</td></tr><tr><td>3</td><td>Chandler Parsons</td><td>423</td><td>15.9</td><td>8121.6</td></tr><tr><td>4</td><td>Dwight Howard</td><td>393</td><td>19.2</td><td>5423.4</td></tr><tr><td>5</td><td>Omri Casspi</td><td>217</td><td>13.8</td><td>2538.9</td></tr><tr><td>6</td><td>Patrick Beverley</td><td>231</td><td>11.7</td><td>2425.5</td></tr><tr><td>7</td><td>Francisco Garcia</td><td>209</td><td>10.5</td><td>2048.2</td></tr><tr><td>8</td><td>Omer Asik</td><td>186</td><td>9.8</td><td>2436.6</td></tr><tr><td>9</td><td>Terrence Jones</td><td>150</td><td>13.1</td><td>3270</td></tr><tr><td>10</td><td>Donatas Motiejunas</td><td>25</td><td>21.8</td><td>890</td></tr><tr><td>11</td><td>Greg Smith</td><td>17</td><td>35.6</td><td>74.8</td></tr><tr><td>12</td><td>Ronnie Brewer</td><td>16</td><td>4.4</td><td>124.8</td></tr><tr><td>13</td><td>Aaron Brooks</td><td>79</td><td>7.8</td><td>0</td></tr><tr><td>Total</td><td>&nbsp;</td><td>2716</td><td>&nbsp;</td><td>40950</td></tr></table>

Team PER = 40950/2716 = 15.08

 

^ dude I'm traveling on yet another biz trip so can't run the stats that I say are easier than that mathematical treatise you wrote above

And yes I do know math. I make my living that still requires a high level. I'm going to go off on a limb and say I'm better at math than Hollinger.

I like your stats. It brought the rockets back down to earth at 15 vs the OP. But when I have time, or maybe someone else can, can someone run the rockets stats through Hollingers PER formula as if the rockets were a single 48 mpg player

Cheers for the work. It's cool