I thought I'd post game flow and box score data from PopcornMachine.net in the thread. For those that don't know, that website charts how the scoring margin changes throughout the game with various lineups on the floor. It's a nice way to visualize which players worked well together in a given game. It also provides an expanded box score, which provides stats for each players in each of their "stints" on the court. How did Yao play in the first quarter versus the fourth quarter? Those types of questions you can begin to answer with it. So I'm just going to post them in this thread for each game this year. Game 1: Memphis @ Houston 10/29 Game 1 Gameflow Game 1 Box score Game 1 Story (Rockets.com)
Thanks durvasa (saved me from going to find the link). At first glance it looks like the worst lineup on the floor in this game would be the Ron, Luther, Barry, Chuck, Landry group. But that is also the same group that gained us the lead back in the fourth. I think the last two minutes of the third quarter and first couple minutes of the 4th would have been better to take Yao out. Those were his lease productive minutes. It may have been better to leave scola in for those few minutes and take yao out. I just love to see the +/- combinations. Something also for people to look at, is who was matched up for the griz with each line up. Edit: For example = Taking Darko out in the third to replace him with Arthur seems to be the spark that lifted the Griz to take a lead.
Check out this page... http://popcornmachine.net/help.html it will give you all the stats and where they come from.
that's really cool. thanks for posting it up. kinda shocking to see that it was mostly our starters responsible for that awful third quarter.
It stands for HelpValue: Reb + Ast + Stl + Blk - TOV Basically, all the box score stats that don't involve directly scoring the ball. You can split up the formula for NBA EFF into: EFF = ScoringStats + HelpValue
hv is help value. It's a metric that combines positive stats and subtracts negative ones (something like points+rebs+assts-tos). Popcornmachine.net is a great site, thanks for introducing it to more fans, durvasa. Our bench team was pretty weak last night. But Ron looked great.
arthur looks like he could be a beast. he looked a lot bigger in person than i remember thinking he was. i agree on the yao comment. seemed like he was having a lot of energy problems late in his stints on the court as well. not boxing out, getting beaten to 50/50 balls, that sort of thing. just goes to show you i guess, that a good set of stats only tells you what you can already see when watching the game. haha.
I would say that it confirms what you can see, or it points you towards things you can go back and see.
The problem with these sorts of analysis are two fold. Statistically - the sample is way too small to really establish any kind of true trend, and the other problem is that it doesn't take into account the opposing team actions. That's why a human eye is still the best judge and the analysis is still behind. Because there are too many statistical variations (open shots missed, a player have a bad or a good game on either team, the way the game is being officiated, etc - that can throw off the analysis) and it doesn't take into account match-ups either. Some players match up better against others. It's a pretty complex set of permutations, one that I don't think analysis can really guide too much.
yeah that's pretty much what i meant. heh. there are so many cool ways to capture basketball stats. someday if i ever get a ton of free time i'd like to do a stats crunching website like that one. just seems like it would be a lot of fun.
but like durvasa said, sometimes taking a look at the statistical results will point your analysis (visual or otherwise) in the right direction. so you spend time asking questions like "how did darrell arthur manage to get 15 rebounds" and what was going on from the rockets' perspective that allowed it to happen? even within a small sample size like this, the "outliers" are still really interesting . . .
The best statistics are the ones that are developed to quantify what a person sees on the court. This particular statical forum is to try and create logical ways to see a snippet of the game combinations through the data available in a box score (and the play by play). As far as this being too small of a sample, the objective of this site isn't to see an overall season, it's to see the on court relationships in one particular game without having to watch the game again Analysis is just a predictor, the problem is people who are skeptical of statistics don't understand that one stat isn't going to be the end all be all of statistics. It's a combination of what you can see quantified in some way.
You agree that its more descriptive than simply looking at the boxscore, right? Obviously, you can't draw strong conclusions from a single game. Even if you're just watching it on TV. This is just a way to capture and record some of the stuff you might observe (or, somehow, didn't observe). It's not a full-fledged analysis that answers all questions. As for not taking into account opposing team actions -- well, not completely. It does tell you who were the opposing lineups faced and their box score statistics (i.e. what they did, as recorded in the boxscore). That's certainly a start, don't you think?
I know the lineup of Artest, Barry, Head, Landry, and Hayes, got us the lead back in the 4th, but they also gave it up and played quite poorly while they were in together. Ron was the only real threat in that lineup and as a result, you saw him ball stopping and attacking the basket from the wing. Artest is typically a ball stopper, but when he was in with the "regulars" the offense was more fluid. I agree that some of the ball stopping with Artest and Tracy is problematic, but none more so than when T-Mac is in with Artest and the rest of the regulars. That's when it becomes a major problem. Ball stopping with Barry, Head, Landy, and Hayes is practically a necessary evil.
I would say that's the opposite of what should happen. When that group is on the floor every one of them should be moving constantly. Ball stopping was the problem with that line up. We all know that none of those guys are going to take someone "one on one" so the only way for them to get any offense is for the ball and bodies to move on that end of the floor. They were all standing around watching Ron. It was like last season all over again when everyone stood around watching Tmac. Since the ball was stopping at the top of the three point arc and there was no movement everything bogged down. If they were cutting and setting screens as the offense is suppose to, then they could have gotten much better looks at the basket.
I'm not sure I agree with that, but maybe I'm misunderstanding. None of them are great penetrators or ISO players, but I think Barry and Head move better without the ball than any other player on the team. Both of those guys are great floor spacers as well, and Barry knows how to pass. Landry, in my opinion, can be a very good high post player, as well as being a good finisher at the basket and offensive rebounder. Hayes is the hustle guy who'll set great screens and chase after the offensive boards. It's modest, but there's some talent there. I don't think they're necessarily ill suited for Adelman's motion offense.
durvasa you and I think so similarly sometimes!! Honestly I think the biggest part of the problem with that lineup last night was head instead of brooks. Though we played with that line up in the last preseason game in Sacramento and they were much more efficient than last night.
In and offense like Adelman's, movement is necessary. There are times in every game, that for whatever reason, the guys just stand around. In those instances, the guys on the floor aren't playing at their optimal performance. I think we can all agree on that. The problem becomes the situations where there is no real facilitator on the floor (and maybe that's the route I should have gone in my first post) to initiate any movement. This situation happened with the lineup I reference. You are right, Barry is a good passer and like I said, I would feel comfortable (outside of Artest) with Barry on the ball. However, that also takes away another good asset of Barry's - his shooting. As for Landry and Hayes, I think we can all agree that they are better off receiving the ball. Regarding Luther, I don't feel comfortable with him initiating anything. Catch the ball and shoot it. Don't do anything else. With all those things said, I feel much better about Ron dominating the ball with that lineup. He can go ISO or at least penetrate and pass out to the now stationary players, two of which excel at spot shooting. I think players often play down or up to each other, much like the concept with teams, i.e. good teams play down to their competition and vice versa. That's why you see guys like Head and Hayes for example that come off the bench and intersperse with the starters. They compliment the starters if you will where as if they played with each other, they might have the deer in the head lights look.
