What does that even mean? Almost every win the Warriors have gotten this year has pretty much been directly and discernibly because of Monta Ellis. Three nights ago, he had 33 points on 13-27 FG and 10 assists as GS beat the Suns in a close one 132-127. Last night, he had 37 points on 15-26 FG as he led the Warriors over the Celtics 103-99, sinking two clutch free-throws to ice it at the end. He has pretty much been the only reason GS has 9 wins. It's not his fault he has had very little help, with no star players and a ton of injuries. Even with the 9-21 record, they have had at least half of those losses stay close until the very end. He's played exceptionally well given his circumstances. Do you think he asks to play 41+ minutes a night? Many players in the league could never handle that. Their "PER" numbers would take a big hit (how sad). He's carrying the team on his shoulders right now, always guarding the opposing team's best player even though he's 6'3". Don Nelson has him guarding guys like Kobe and Arenas. He even locked down Brandon Roy himself (Roy still scored 17 but on 6-17 shooting) as the Warriors thumped the Blazers 108-94. Here's an excerpt from the recap: Honestly, I don't see that there's much more he could possibly do. He's all-star caliber, that's for sure. And his team seems to be coming together and playing better, now that they aren't hobbled with a 6-man rotation. He may actually be able to take a rest once in a while.
Yes, Monta's played wonderful, but that's not what an All-Star should be. I'm a homer and I'd be slightly embarrassed to see him go. Why? Because the team stinks and the product is awful, even with the two game winning streak. Monta suffers from the "Rich Kelley/Lloyd B. Free" syndrome, no one really cares about what you do if your team stinks, unless you are Pete Maravich. The thinking goes that since the team stinks, the rebounds and points don't matter. Its about winning, and I tend to agree. I would rather see Deron Williams make it than Monta, really.
That's what I said in my first post. The reason why he won't get voted in is because he is on a terrible team and is kind of unknown. It doesn't change the fact that he is playing at an all-star caliber level. And yes, I agree, Deron Williams is a better player than Monta, he's playing great and the Jazz have a winning record. I would vote for Deron over Monta.
Replace Ellis with a D-league guard, and use a combination of Curry, Azabuike, and Watson to make up for his absence in the rotation. How many less wins would you say the Warriors have?
3 role players don't make up for one star player (all-star, superstar, key player, whatever you want to call it). You're totally overvaluing our role players and undervaluing Ellis'. Even if those players combine for a similar stat-line, you're neglecting the impact a star player has in a game (how the other team adjusts to him, how it opens up things for his teammates, etc.). While it'd be nice to get some wins, the Warriors need to develop an identity and Ellis is obviously a part of it.
Azibuike only played 7 games before going out for the year anyway. And Steph Curry plays 31 minutes a night, almost all the time alongside Ellis. Those two wouldn't replace him. It would basically be your so-called D-League guard, Anthony Morrow, and CJ Watson. Let's see... they're 9-22... without Ellis, they'd probably be something like 3-28 or 4-27. I have a hunch you haven't seen many of the Warriors' games this year (I could be wrong). Anyone who has watched the Warriors will tell you just how important Monta has been to this team. It's not only his quantifiable numbers -- it's his qualitative impact on the game. He is like a laser-beam, all game, for 40+ minutes, every single night, getting tips, deflections, steals, playing aggressive defense on bigger guards, handling the ball for the majority of the game, initiating the offense for others, driving into the teeth of the defense to allow others to clean it up -- many of his drives result in collapsing the defense, breaking down the interior -- he doesn't always get a point or assist, but he wreaks havok on the defense's structure and allows others to take advantage of it. For example, he drives the lane constantly, every game, but unlike some point guards who are just being pesty, he is driving to attack the rim and finish over 7-footers. Many times, he draws 3 defenders and goes up over them all. He doesn't always score, but guys like Anthony Randolph and even Corey Maggette have an easy time going up for tip-ins and putbacks off it. Also, the 3-pt shooters, Morrow, Curry, Watson, Radmonovic -- they all benefit because the defense usually sucks in to deal with Monta. He can dish it out, and it will then swing again to the now wide-open man for an easy 3. Monta won't get the assist, but he's the main reason why they play was so successful. Without Monta, this team would be absolutely lost. They'd have no threat off the dribble, no inside-out game, and no playmaker. The outside shooters would never get open. Corey Maggette would be taking 30+ shots a night. And this team would be atrocious to watch (yes, even more-so than right now). I'm not sure what the debate is really about here? No one is saying Monta Ellis is the best player in the NBA. Frankly, I probably wouldn't vote for him for the all-star team myself. But it should be obvious, to anyone who knows the game and watches this team, that Monta Ellis is playing at an all-star caliber level this year.
Your hunch is correct. I have not watch the Warriors play, outside of their games against the Rockets. As you describe him, he indeed sounds like he's been having a terrific season. I guess the numbers sell him short. I can say that, typically, very good players on bad teams will have decent On/Off +/- figures (meaning, the team doesn't get killed as bad with them on the floor than with them off the floor). The argument I've read in defense of Ellis that he doesn't leave the floor except in garbage time, so the performance of the team with him off the court is inflated. Here's the actual data through 12/27 (if you're interested): When the margin is within 5 points: Code: ORTG(#poss) DRTG(#poss) NET Ellis On 106.5(1329) 106.9(1276) -0.4 Ellis Off 124.7(77) 115.1(73) +9.6 Where the margin is within 10 points: Code: ORTG(#poss) DRTG(#poss) NET Ellis On 105.2(1950) 108.6(1905) -3.3 Ellis Off 126.4(163) 121.5(163) +4.9 So its true that Ellis plays in a very high percentage of the possessions when the game is close (about 95% of them when its within 5 points, 88% of them when its within 10 points). But, the story is the same. Better net performance with Ellis off the floor than on the floor when the game is close. Granted, there is apparently a steep dropoff on the defensive end. In your opinion, is Ellis more valuable for his defense or offense? Where do you think he has a bigger positive impact? It looks like it might be on the defensive end.
Frankly, you just need to watch him play a few games. All those stats you're throwing around don't mean anything.
It doesn't mean anything that the Warriors outscore their opponents when games are close with Ellis off the floor? That seems relevant to me. If nothing else it is an interesting puzzle. Since you watch the Warriors regularly, you tell me what's going on there. But I can't just ignore the fact that the Warriors offensive efficiency is over 15 points better with Ellis off the court. That just leaps off the stat sheet and demands being addressed. We're talking about only a few hundred possessions, which perhaps covers only a small handful of games. Maybe the vast majority of those "off-court" possessions were against really, really bad defensive lineups. That could be the explanation.
How do you know the games are close with Ellis off the floor based on the stats you present? That's not what those stats say. Ellis is a crunch time player, so he's always on the court near the end of close games. Second, you aren't giving his on court time and off court time fair treatment. His on court time should be given more weight than his off court time because Ellis is on the court most of the time and the difference is close to twelve-to-one based on the number of possessions. If the team does not do well, then Ellis does not do well. You're also giving equal weight to the second team players vs the opponents secondary players when Ellis is off the court since Ellis has been on the court most of the time. It does not necessarily mean that he's not an all-star player. You have to look at his other stats to properly evaluate how he is doing. I don't like using the stats you present because those stats are misleading. For example, if a player does not do much offensively like Thabo Sefalosha, but still is on the court with his teammates who are scoring, then he gets the same value as his teammates. In Ellis' case, he's pulling up the offensive value of his teammates when he's on the court. Of course, if the team does not do well, the he doesn't do well either. If you look at Ellis' other stats, then you find that he is not as efficient a player as other players like Kobe. I like to use PER when comparing players. I just looked it up, and he's not in Kobe's league, but he's doing a skosh better than what Brandon Roy is putting up this season. If Monta became more efficient and used better shot selection when he's struggling (or improve his FT or 3-pt shooting), then he would rank well with the other all-stars at his position. In a lot of cases, Monta would beat them out because of the other things he excels in. EDIT: I think we all use stats to some degree when selecting our all-stars. It's not like we can watch most of the all-star candidates all the time. It's also a popularity contest and based on fan's perception than a more professional opinion. It's not a perfect system, but I rather vote for the all-stars than see some lame brain coach/player voting or stupe sports writer voting. LOL.
Well, in regards to the data you mention, I'd say two things: first, it's a small sample size, and by the end of the season, those numbers could drastically change. Second, usually, when Nelson does his rotations, he leaves Ellis in to play with the back-ups. The other 4 starters usually end up sitting by early 2nd quarter, and Ellis stays in to play with the back-ups until half-time. When Ellis does sit, then Nelson will always put several of the starters back in to have some leadership. So that may affect things, a bit (maybe). Regardless, I really wouldn't invest a lot in those numbers right now, with only 31 games having gone by. Ellis has been equally important to the Warriors few wins offensively as he has been defensively. He's been a leader, too, and never gets out-worked. He always sprints back on defense, plays defense the right way, and rarely argues with the referee, even though he gets hammered by 7 footers all day. Trust me, if you asked any of his teammates, they'd tell you how valuable he is to them.
It shows points scored and points allowed per possession when the score is within 5 points. If the score is within 5 points, I would consider that a "close" game. I don't know what kind of weight you're talking about. He is a net negative in his court time in close games. If you want that to be given more weight, that would only hurt him in this evaluation. And, yes, it is true that if the team doesn't do well, he doesn't do well. But the question I have is why the team seemingly does very well when he is off the floor. As I said before, if the vast majority of those possessions span only a few games, we could write it off as a fluke. But if those possessions are from several games against many different types of lineups, it could indicate a meaningful trend that the Warriors are doing something better with Ellis not on the court. An adjusted +/- analysis would adjust for opponent strength. Such stats are available at basketballvalue.com, and it shows that when adjusting for this Ellis still does not rate well in terms of impact on the scoreboard. But if a player is hurting his team by not scoring, then his team would play better without him, right? Similarly, if a player is carrying his team by scoring the ball, then they would play better with him. Ultimately, shouldn't we care about how the team performs with the player on the court? Isn't that what really matters? If we agree that is the case, then its a matter of figuring out a way to best capture that impact. Its not easy to do, and the stats I presented aren't perfect by any means. But I do believe we should be judging a player's impact along those lines. And I'm not trying to make a case against Ellis; I'm just looking for an explanation.
I know that, but I'm talking about your interpretation of net negative. Let's just stick to close games. It's not because of Ellis being on the floor that causes it. It's caused by the Warriors sucking at the end of games as a team. You can't just point the finger at Ellis, but I suppose a lot of people will do that since he's our best player and the team is losing. If the Warriors won games, then the net negative would be a positive. At least we agree on that. Even with a net positive, it does not necessarily mean that Ellis won the games or was a main contributor to winning. It just shows that the team did well and that Ellis was a part of that team. It would mean that that particular combination of players played well and would be a way to compare how well players do in playing together. I said fair treatment and you're not giving Ellis that. You're arguing based on a limited interpretation of the stats. If you were more fair in your evaluation, then you would realize that weight would come into play based on the # of possessions when using these stats and comparing player combinations. You are talking about the net positive side now. It does not mean the team is doing better because Ellis is off the floor. It just means that for seventy something possessions, the team did well and Ellis just happened to be off the floor. Again these stats reflect the team more and not the individual player. Well, I won't comment because you need to provide more specific examples of what you are referring to. Not necessarily. I gave the example of Sefalosha. He does not do much scoring, but he's a defensive player and helps the team that way, so he starts and gets some minutes on his team. Again, your stats do not show that a player would be hurting his team by not scoring. I did give you an explanation for how well Monta Ellis is doing by evaluating a player using other stats such as PER. Your stat would be useful when evaluating how Curry and Ellis play together. Why I bring that up is because the Warriors do not want two small men on the court on defense for a majority of the time. If that's a net negative and the team is not performing well, then one of the things we need to look at would be changing the lineup. Of course, if the Curry-Ellis combo improved the offense side of the ball to be greater than the points it gave up, then we would have a net positive. Still, we would have to evaluate all the possible combinations that the Curry-Ellis duo had with the other players. It may not be just Curry-Ellis. I like looking at metrics for evaluating basketball, but the game is not made for it because trying to quantify and evaluate teamwork makes it difficult. I think your +/- numbers give us a snapshot of how a certain combination of players did in the past. However, I have not seen it used to set the lineup during crunch time of specific games. I doubt any coach uses metrics to set his lineup, but uses what combos worked during that particular game and goes with someone who has the hot hand that game.
Some really good dialogue here. One thing to remember, is that the Warriors have been playing without a big man (Wright, Beidrins, and Turiaf have been out the majority of the season). That being said, their only chance of staying in games has been Ellis taking over games. Other teams have been eating the Warriors up on the boards, and still Monta has been able to score and keep them in games that they had no business being in. I'm not the biggest supporter of Monta, in that he turns the ball over WAY too much...However, like I mentioned before, he's been doing this without big men, so Monta's been forced to try to do too much. We'll get a better idea if he's an All-Star by this time next year. I'd still hope that he gets some votes, my "young guy" vote goes to Monta, slightly edging out Evans, only because I really, really hate the Kings...but I think Evans deserves it a bit more at this point in the season.
