We all know points differential is correlated with win %. I took the time to log in all teams from last decade with winning % and + points differential into a graph and see where Rockets stand. I then drew an imaginary line (indicated by the yellow) from all the data collected. Teams fell above it, I classified them as an "overachievers" for their accomplishment. Teams fell below it as "underachievers." What this yellow line means is to predict a team record by their points differential. According to our differential of +3.2, we should have a win/loss % of at least .600. If my math is correct, we should have a record around of 39-25 as of now but instead, we are sitting at 34-30. We have underachieved by 4-5 games by our points differential. Yes, we might be inexperienced at close games, but it doesn't justify as the sole reason. OKC 2009, Points Differential +3.5, Win % .610 OKC 2010, Points Differential +3.8, Win % .671 Will this 4-5 games make a difference in the final standing? Most likely not. By looking at the remaining schedule, Rockets will likely to finish with the 6th seed. Even with 5 more wins, we will not be seeded above that. Finally, this is not a hate thread on our coaching staff, as I mentioned many times on other posts, McHale has overachieved with this young and inexperienced team.
On surface, we overachieved. In this graph, we underachieved. We all know Morey is looking for playoff run even before the Harden trade, according to his interviews preseason. So his calculation maybe different from media and average fans.
I brought this up a bit earlier. But to re-hash, the Rockets SHOULD have a better record. You look at our losses, and they're mostly winnable games. Many of them we lost in OT or in the last minute. The question is WHY we keep losing. And the thing is, better coaches don't necessarily mean better close game results. Keith Smart for example is one of the worst coaches in the NBA right now. Thibadeau is one of the best. Sacramento is "overachieving" by 2 games. Chicago 1. My own opinion, our turbulent rotation throughout the season. It's not just about being young, but having such roster turnover from training camp to beginning of the season to the trade deadline. Next season, if we still have the same problem, is when I'd start looking at McHale.
the way we're assembled, even before the harden trade, I thought we would be hovering around the 9th seed possibly the 8th seed. after the harden trade, yeah we should be 12 games over .500. We've underachieved.
Good work - thanks for putting this together. I know the advanced stats guys love using point differential as the primary factor in projecting wins - because it's been the most reliable over the years. But I do think in this instant case, the context is important. Not merely the youth of this team, but also the relative inexperience and the Feb. deadline trade shake-up cost this team some games. If this team is only underachieving by 4-5 games, at least two of those games came after the trade (WAS, PHX) - and at least one more game can be accounted for with injury (Harden in UTA). We can choose to blame youth or McHale's rotations on the other 1 or 2 games this team "should" have won.
Didn't most of the media think that the Rockets wouldn't make the playoffs? Youngest team in the league, with only one real proven player (Harden), and only one remaining player from last year (Parsons) ? I'll say Rockets definitely are achieving currently.
On the season, HOU has outscored opponents by 205 points. In its 3 wins with largest margin of victory (45 pts vs Utah, 33 pts vs DAL, 31 pts vs GSW), HOU has outscored opponents by 109 points. The high mean margin of victory is skewed somewhat by these arguably outlier games (and the lack of outlier blowout losses). Does it mean that HOU is better than its records? Is it simply a matter of the team tending not to slow down and keep "running up the score" in garbage time and is thus relatively meaningless? Or is it evidence of some potential and inconsistency? Hard to say. But I don't think we can simply say that the margin of wins means HOU underachieved.
The yellow line is a correlation. It is not cause and effect. It would be more accurate to say a team with a +3.2 differential should have a win/loss percentage between 0.560 and 0.660. That would put us between 36-28 and 42-22.
The word "underachieved" has a negative connotation. But there is nothing inherently negative about winning less than the expected win% based on point differential. All it means is that we win by more points than we lose. Our point differential in wins is +13.9 and in losses its -9.0. Is that a bad thing?
Seriously, the Rockets have a pretty "feast or famine" record because they rack up a fair amount of blowout wins when teams decide to pack it in (Utah, GSW, Knicks, etc) against a high paced, offensively explosive team, and then unsurprisingly the Rockets themselves fade late in the second game of back to backs to lack of depth/fatigue reslulting in closer losses.
Your key assumption is that points differential is directly related to winning percentage, but i think this is where it is the problem. While I do agree that a healthy point differential should mean that the team is winning more games, I think this doesn't apply fully to the rockets for one reason ---- our up-tempo style of play In my opinion, we have blew out a few teams by huge margins and this has inflated our point differential: November 10th: Hou 96 - Det 82 (+14) November 23rd: Hou 131 - Nyk 103 (+28) December 14th: Hou 101 - Bos 89 (+12) December 17th: Hou 109 - Nyk 96 (+13) December 19th: Hou 125 - Phi 103 (+22) December 22nd: Hou 121 - Mem 96 (+25) December 25th: Hou 120 - Chi 97 (+27) December 31st: Hou 123 - Atl 104 (+19) January 2nd: Hou 104 - No 92 (+12) January 4th: Hou 115 - Mil 101 (+14) January 8th: Hou 125 - Lal 112 (+13) January 25th: Hou 100 - No 82 (+18) January 26th: Hou 119 - Bkn 106 (+13) January 28th: Hou 125 - Uth 80 (+45) February 2nd: Hou 109 - Cha 95 (+14) February 5th: Hou 140 - Gsw 109 (+31) February 8th: Hou 118 - Por 103 (+15) March 3rd: Hou 136 - Dal 103 (+33) The Rockets have won their opponents in these 18 games by an average of 20.45 points. However, they only count as 18 wins out of the 34 (More than half, by the way) we have so far. Thus this leads to a huge point differential but it might not necessarily suggest that we should have more wins. Also you should factor in certain games the rockets have lost by a small margin such as the recent games against the bucks, the wizards and the suns, which wouldn't affect our point differential THAT much, but it still counts as a loss. I know we have lost some games by a huge margin and maybe some of you would argue that it balances out - but I still think that our up-tempo offense has led to point differentials that are unreliable in determining a suitable winning percentage. Hopefully I am understanding your theory right here, if not my apologies. But back to your question, no the rockets are definitely overachieving this year. They have really surpassed all my expectations and its really quite fun to watch the team play despite all the frustrating moments that comes as a result of their inexperience.
i'm not sure the margin of victory has anything to do with over-achieving or underachieving. One troubling sign for me is the number of close games we lose. All i know is that number needs to change for the better. But what would you expect from the youngest team in the league. Seems like we are right where we should be.
Your argument assumes that these large point differentials are unique to this team. It's likely that every team has outliers like this. Here are some more stats: http://www.nba.com/standings/team_record_comparison/conferenceNew_Mrg_Alp.html Look at the column "10 pts or more"...you see that the majority of NBA games are not close. The better teams win more blowouts than lose. I threw these into an excel spreadsheet and wouldn't you know it...win percentage for games decided by 10+ pts had a 98% correlation with margin.
There's only one team with a lower point differential than ours that is ranked higher than us in our conference. We're right about where we should be if you are only going by point differential.
But you're focusing on the wrong thing here. Winning by 40 and winning by 11 still counts as a 'game decided by 10 or more'. One team can win all the games decided by 10 or more but have a lower point differential compared to a team who has lost more games decided by 10 or more. (don't really know if you get what I mean) The point is that, point differential applies to like teams like OKC, but just maybe not us, due to our uptempo offense.
I don't recall media talking about it after the Harden trade. All the NBA previews came out before the trade was made, as I recall. I know that the SI NBA preview.
