I’m 100% with you on this one. That was a great way to break it down. Anyway, back to work.

If ‘Sheed were a math guy he’d say “stat don’t lie”.

I make a living doing charts like this. I applaud the effort to try and use objective analysis to get at the truth. I love it… these are the kinds of posts that make this blog more interesting than, “I just don’t feel good about Travis taking lots of shots. ’Cause my gut says so.”

But here’s the problem.

By virtue of the fact that Outlaw is NOT a starter, he’s going to be playing more in losses and lopsided scores (read: garbage time) because the coaches want to get him some run along with the other non-key players. Every blowout loss is going to feature Outlaw along with some other second-tier players. I expect you’ll find that Martell took more shots in losses (read: garbage time and pseudo-garbage time) and probably Surge and Jack as well. Maybe even Joel.

Jason Quick makes this same problem of ignoring score and game situations when he runs the number of minutes played in the fourth quarters. After all, who cares who plays in garbage time? There are fourth quarters and there are FOURTH QUARTERS you know.

To clean up your stats and make the conclusion more compelling, I suggest re-running your frequency table to include only those games where the difference in points between the teams was, say, 10-and-under or so. (or 15 or whatever… just have a theory) In this way, you know that Outlaw was contributing in critical game situations.

Another—albeit more controversial way to go after this—is to examine FGA and FGM in light of the +/- system. Single games are not good tests… but you should have a season’s worth of data to see if the Blazers tend to outscore opponents the more shots that Outlaw makes. Though again, it’d be nice if you could screen out garbage time somehow.

But I think the kind of frequency chart you’re designing doesn’t is a little facile in describing the reality of the situation. We only care about Trout’s shots while the game’s conclusion is in question.

Despite Phizbin’s comments I think there is some value to what’s posted above.

Rather than equating Travis Outlaw’s increased shot attempts to a decrease in wins, first you must ask yourself the underlying reason for the increase in shot attempts.

Major Reasons for Outlaw shooting 17+ attempts:
1. Brandon Roy is not playing: I believe over half of those games where Outlaw had 17+ FGA were in games that Brandon Roy either didn’t play or played or was pulled due to injury. In this case, the natural conclusion to be drawn is that the Blazers a worse team when Brandon Roy is not in the lineup.

2. Either Roy or Aldridge (our two main FGA’s) are struggling: This is most obvious in the game against Orlando and Toronto. When LaMarcus Aldridge is being neutralized by a better PF, more of the shot attempts go to Outlaw.

3. Travis Outlaw is having a good night: There are occasions where Outlaw shot well and neither of the other two factors were present. In this case, you would prefer that Outlaw take a lot of shots.

I don’t believe there was a game all season where: Outlaw shot 17+ attempts; Outlaw shot under 50% from the floor; Aldridge or Roy didn’t struggle; and Roy played in the game.

Outlaw’s increased shooting doesn’t impact the team losing… it is a byproduct of a larger factor that inhibits the team’s ability to win.

At first glance, these stats look like they speak for themselves. But this doesn’t necessarily tell the whole picture. I compare this to the +/- category. For example, there may have been times when Nate may have thrown in the towel (down by ~20) and gave the green lite to Trout late in the 4th quarter. Not saying this always happened, but it’s not out of the realm of possibility.

I love numbers. I agree with everyone’s critique of the “research design” that Ben used. Since Travis FG% was not randomly assigned, we cannot be sure of the causal effect of his shot attempts on wins. That being said, I think that your conclusion at the end is still valid. The burden of proof is on Travis, given that there is no evidence that it helps to have him shoot more.

I’d like to point out a slightly more elementary argument: 43% is lower than the average FG% on the team (and the disparity is wider for EFG%). Unless being able to shoot more frequently increased Outlaw’s shooting percentage, his shooting more cannot increase the Blazers chances of winning.

Now I’m pretty sure it’s a bad idea and he needs to just accept his role unless he can start shooting a 45% or better. Thanks for taking the time to do all the math and graphs.

As long as they are all drives to the rim.

I especially wanted to look at Nate’s claim that Travis should post up more to become an unstoppable “dirty” player (his words, not mine) who plays for winning the game and not for the outside jump shot (see Quick part I, II). I’m sure NBA teams use even more sophisticated shooting spot charts to find weaknesses, but this is what e.g. NBA Live employs. I think it’s more visual than shot charts with X and O positions for each shot, and those are usually just game by game (I looked at a few mentioned by Salem Stephen above where he had more than 17 FGAs and then gave up). Here red is “hot”, greyish-brown is “medium”, blue is “cold” (relatively speaking). Also given are the number or shots (made and attempted) as well as the percentage from each zone.

Fig. 1 Travis Outlaw Shot Zones in the 2006-07 Regular Season

Fig. 2 Travis Outlaw Shot Zones in the 2007-08 Regular Season.
Notice the improvement in “cold zones” especially on the left wing, and the overall increase in attempts.

Fig. 3 As a comparison with our other SF: Martell Webster Shot Zones in the 2007-08 Regular Season

—-
Fig. 4 out of competition: Why Sergio needs all the time with a shooting coach he can get

I’m curious, if Travis is taking a ton of shots, I’m thinking that would mean he was playing more minutes, and if he’s playing more minutes, I assume that means someone else is hurt. Could it be that Travis’s extra shooting and losses happen to coincide when say, Lamarcus or Brandon is injured and Travis is asked to take on more of the offensive load? Is that the case? What was the roster like on those big shooting nights?

Very informative, and nicely done. Non biased as well. I <3 you.

Don’t Trade Him.

Dave. I think your charts provide good info, but are incomplete. Shots per minute would be a better indicator than shots per game. Playing time is an issue as is quality time. Was he playing a lot during a blowout? During a close game. There is just more to it.

Because I’m one who doesn’t really like to assign shots before games. I think that we should take adv. of matchups, and take them in the flow of the game.

I don’t think Outlaw should try and get his shots up every game, because I don’t think that it makes us any better.

I’m really hoping that he gets better at playing withing the flow of the game, and passing, and not just going 1on1 every time he gets it. If he rebounds and plays D, and plays within the flow of the game, his shots will come, and the team will be better for it.

Thanks for giving us the data.

I think Coach would probably agree with you that Outlaw needs to earn his increased scoring role. He’s certainly on the way there—close to all the way there if you just consider the late game—but as long as he “falls in love with his jumper” as Quick quoted him saying I think he’ll have to keep proving himself. Also everybody on this team will have to sacrifice some shots and some of their offensive ability in order to play nice together. The team priority will be making that happen and fitting these guys together. Here’s hoping Travis can be happy with what he gets and earn as much as he can.

—Dave

If he gets 15 plus shots a game, Nate should be fired.