If the season started tomorrow...

Heltonfan · 11-22-2006, 12:18 PM

Here's how I see the NL West right now:

LA 86-76
SD 85-77
AZ 80-82
COL 79-83
SF 69-93

Obviously, there are some very incomplete teams in this division; the Padres, for instance, project to those 85 wins despite having only three starting pitchers (I assume replacement-level performance from the empty slots) and no second baseman. So merely by filling those gaps, they'll get up around 90 wins.

Anyway, the point is that, thanks in large part to the Dodgers' nice gesture of throwing away more than $60 million on two players that they don't need, we're actually not that far away from serious contention. A couple nice additions, and we'll look like a .500 team going into the season... and that's despite the fact that I'm projecting substantial regression from Atkins and Holliday, and I'm not expecting anything at all out of Tulo. So there's certainly room for the projections to be beaten.

Roxpert · 11-22-2006, 05:07 PM

To what would you attribute a 3-game improvement for the Rockies, Heltonfan? Since you look for regression from Atkins and Holliday, where will the slack be picked up? Clearly, it will be hard for the Rockies to top their pitching success of 2006 (not so much due to out-performing projections, but more due to the superb health of the staff last season).

My guess is that the Rockies aren't improving in your projection model as much as the competition is getting worse, allowing the Rockies to tack on a few more wins. Unless the Rockies infuse the team with more depth in the bullpen and talent on the bench and in CF, I don't see ANY improvement in the performance of the team, and in fact I see them getting worse. A lot more went right with the 2006 Rox than went wrong.

Heltonfan · 11-22-2006, 05:36 PM

Quote:

Originally Posted by Roxpert

Since you look for regression from Atkins and Holliday, where will the slack be picked up?

C, 1B, and SS. Iannetta projects pretty well, and perhaps just as importantly, we don't have Danny Ardoin around to kill the offense any more. Helton should bounce back somewhat (I know you probably disagree with this one... but the multi-year weighted average projection model is what it is because it works). And if we get merely replacement-level performance out of our shortstops this year, that will be a huge improvement.

And yes, you're right that some of the Rockies' apparent improvement is really just the competition getting worse. But that doesn't make it any less significant.

Finally, as far as the projections are concerned, 79 wins isn't a 3-game improvement from 2006, it's a 3-game dropoff. My system doesn't care that the Rockies underperformed their component stats by six games last year.

Roxpert · 11-22-2006, 06:01 PM

Quote:

Originally Posted by Heltonfan

Finally, as far as the projections are concerned, 79 wins isn't a 3-game improvement from 2006, it's a 3-game dropoff. My system doesn't care that the Rockies underperformed their component stats by six games last year.

Uh-oh. If your system is resulting in a 3-game dropoff, masked by the actual record, then we can conclude that Hurdle will get about 75 wins from this group of players. In the four full seasons that Hurdle has managed (not counting 2002 when he inherited a 6-16 team), the Rockies have played a total of 17 games worse than their Pythagorean record, and has been amazingly consistent in under-achieving vs. RS/RA. The team has been between 3 and 5 games worse the past four years.

So, it's safe to assume they will be about 4 games worse than a projection system such as your's suggests. The NUMBERS may point to 79 wins with your methodology, but the TRACK RECORD of the field management suggests this is more likely a 75-win projection.

Heltonfan · 11-22-2006, 06:43 PM

Quote:

Originally Posted by Roxpert

In the four full seasons that Hurdle has managed (not counting 2002 when he inherited a 6-16 team), the Rockies have played a total of 17 games worse than their Pythagorean record,

Well, it's awfully convenient not to count 2002, isn't it? The 2002 Hurdle-managed Rockies beat their Pythag by 5 games. So now the difference is down to 12 games in 5 years, and that's still making the (as far as I know, unsupported) assumption that variance from Pythag has predictive value. It's a real reach to assume that we'll underperform Pythag by more than a game or two, and I think that neutral Pythag luck is the best assumption to make.

Roxpert · 11-22-2006, 07:30 PM

I didn't exclude 2002 for "convenience", Heltonfan. I simply am not somewhere that allows me to look at the data for the 140 games that Hurdle managed in 2002. So I only see that the team played 3 games better than Pythagorus in 2002, and can't see that it was 5 games better during Hurdle's reign, but I'll take your word for it.

Assuming that's right, the following is the Hurdle record vs. Pythag....

2002: +5
2003: -4
2004: -5
2005: -3
2006: -5

See any pattern there? If anything, I think 2006 UNDER-reports the number of wins that Hurdle cost his team, and I think you agreed that he was very costly last season.

If you have the same inept and incompetent field management, how can you NOT incorporate that into an overall projection of their record? Strictly on a compilation of the numbers, your 79 wins is something I cannot dispute. But, based on Pythagorean AND empirical evidence of Hurdle's self-destructive managerial tendencies, I can and do discount such a projection by a handful of games. It may be subjective to give a haircut to your 79 win projection, but I believe it's proper.

Heltonfan · 11-22-2006, 07:40 PM

Quote:

Originally Posted by Roxpert

See any pattern there?

Sure, there's a pattern. Is it meaningful? Is there any evidence that the trend is more meaningful than the weighted average, regressed somewhat because it's insane to assume that the Pythag variance is completely due to the manager? I think not.

I think it's perfectly legitimate to knock the projection down by a couple games to account for Hurdle's poor use of the roster, though. No argument there. But the argument shouldn't come from the Pythag variance.

Roxpert · 11-22-2006, 07:53 PM

I show the Pythag. results not to literally say that it proves that Hurdle costs our team 4 wins per season, but only as further illustrative support that indicates that this team rarely seems to live up to the implied wins that RS/RA would suggest.

Still, I think it's at least plausible to subjectively assume that if your system comes up with 79 wins, the field management will take the raw material and mold it into, say, a 75 win team. In other words, your quant-method is a good one, but inevitably incomplete due to its inability to account for the "Hurdle" factor.

Actually, I think the Hurdle factor is even MORE costly than the Pythag. numbers suggest. I've felt for years that he sub-optimizes his rosters, and wins 5 to 10 fewer games per season than the raw talent given him would allow. I haven't seen any improvement in his managing decisions either in recent years. Granted, this is ALL subjective, but Pythag. lends further support to my subjective belief from the empircal evidence that the Rockies will usually not quite live up to their talent level on paper.

Having said that, we actually did exceed our talent level in 2006. It was an outlier season even though Clint Hurdle cost us several wins. I don't think that bodes well for 2007, by the way.

wolf213 · 11-23-2006, 12:48 AM

I actually think 79 wins is right on, I don't like Hurdle and do think he cost us a ton of games. I would love to see him manage in the national spotlight like the playoffs so the whole world can see how bad he is.

I predict drop offs for Atkins and Holliday, but not by much, and they also can very easily stay the same. I also expect improvement from Hawpe and Baker, and if Tulo is even league average, then the SS position would be amazingly improved.

indianadrew · 11-23-2006, 09:36 AM

isn't it a manrule that we can claim it was a rebuilding season? we still had a better season than 2005 though, right? i think that the most disapointing thing was that we were in alot of the games we lost. it does not seem to me that it is going to take too much to improve from last year. GO HORSE!

madness · 11-23-2006, 10:19 AM

Quote:

Originally Posted by Roxpert

.....Still, I think it's at least plausible to subjectively assume that if your system comes up with 79 wins, the field management will take the raw material and mold it into, say, a 75 win team. In other words, your quant-method is a good one, but inevitably incomplete due to its inability to account for the "Hurdle" factor.............

I may be wrong here (and attempting to speak to someone else's motivations and objectives is always dangerous) but I think that Heltonfan was providing a quantitative projection of 2007 records to serve as a baseline for a complete projection (one which would factor in qualitative values as well as factors which are not statistically projectable).

This may be one of those arguments that you could have with someone for hours on end and then when you finally give it up and hit the pub later and ask them casually if they think that the Rockies will really win 79 games or if they would put thier own money on that they will respond "Hell no, I was just defending the math. I bet that dimwit Hurdle brings back Morgan Freeman to pitch the 8th inning so that he has a "veteran reliever" and costs the Rockies a dozen games" or something similar.

Cliffs: It is my impression that his "quant-method" is intentionally 'incomplete' and wasn't intended as a final projection that included all factors (including those that defy statistical projection).

rockieprogress · 11-23-2006, 11:35 PM

If there is any improvement, and I tend to agree that the 3 areas that HF thinks will improve are the most likely, how can they not improve with the poor showing with RISP? Just avg in that area, and they should be able to gain the 3 games.

With this off-season playing out as it has, next year is going to be very interesting.

wolf213 · 11-28-2006, 12:46 AM

Here is my 2 cents.

1. D-Backs
2. Padres
3. Dodgers
4. Rockies
5. Giants

I think we can all agree that the Giants arn't going to be very good next year after loosing their Ace, and 2 of their sluggers.

D-Backs are looking good in my opinion, they have smart GM who made some nice trades, and they have some good depth in the system

Padres are standing pat so far, but I expect them to get a pitcher and maybe another bat. I have learned from season's past never to discount them

Dodgers are making dumb move after dumb move. Pierre takes of for Drew? Wolf for 8 Million? Nomar? They have plenty of depth but I'm crossing my fingers they do something really stupid and give up a ton for Manny.

Rockies would be ranked higher however Hurdle is still the manager of this team. Even though I have them in 4th I think it will be a very competitive 4th, maybe with a .500 recored.

11-22-2006, 12:18 PM	#1 (permalink)
Heltonfan Moderator Join Date: Nov 2006 Location: Yerevan, Armenia Posts: 855	If the season started tomorrow... Here's how I see the NL West right now: LA 86-76 SD 85-77 AZ 80-82 COL 79-83 SF 69-93 Obviously, there are some very incomplete teams in this division; the Padres, for instance, project to those 85 wins despite having only three starting pitchers (I assume replacement-level performance from the empty slots) and no second baseman. So merely by filling those gaps, they'll get up around 90 wins. Anyway, the point is that, thanks in large part to the Dodgers' nice gesture of throwing away more than $60 million on two players that they don't need, we're actually not that far away from serious contention. A couple nice additions, and we'll look like a .500 team going into the season... and that's despite the fact that I'm projecting substantial regression from Atkins and Holliday, and I'm not expecting anything at all out of Tulo. So there's certainly room for the projections to be beaten.

11-22-2006, 07:53 PM	#8 (permalink)
Roxpert Veteran Member Join Date: Nov 2006 Location: Denver Posts: 1,059	I show the Pythag. results not to literally say that it proves that Hurdle costs our team 4 wins per season, but only as further illustrative support that indicates that this team rarely seems to live up to the implied wins that RS/RA would suggest. Still, I think it's at least plausible to subjectively assume that if your system comes up with 79 wins, the field management will take the raw material and mold it into, say, a 75 win team. In other words, your quant-method is a good one, but inevitably incomplete due to its inability to account for the "Hurdle" factor. Actually, I think the Hurdle factor is even MORE costly than the Pythag. numbers suggest. I've felt for years that he sub-optimizes his rosters, and wins 5 to 10 fewer games per season than the raw talent given him would allow. I haven't seen any improvement in his managing decisions either in recent years. Granted, this is ALL subjective, but Pythag. lends further support to my subjective belief from the empircal evidence that the Rockies will usually not quite live up to their talent level on paper. Having said that, we actually did exceed our talent level in 2006. It was an outlier season even though Clint Hurdle cost us several wins. I don't think that bodes well for 2007, by the way. Last edited by Roxpert; 11-22-2006 at 07:57 PM.

11-22-2006, 05:07 PM	#2 (permalink)
Roxpert Veteran Member Join Date: Nov 2006 Location: Denver Posts: 1,059	To what would you attribute a 3-game improvement for the Rockies, Heltonfan? Since you look for regression from Atkins and Holliday, where will the slack be picked up? Clearly, it will be hard for the Rockies to top their pitching success of 2006 (not so much due to out-performing projections, but more due to the superb health of the staff last season). My guess is that the Rockies aren't improving in your projection model as much as the competition is getting worse, allowing the Rockies to tack on a few more wins. Unless the Rockies infuse the team with more depth in the bullpen and talent on the bench and in CF, I don't see ANY improvement in the performance of the team, and in fact I see them getting worse. A lot more went right with the 2006 Rox than went wrong.

11-22-2006, 07:30 PM	#6 (permalink)
Roxpert Veteran Member Join Date: Nov 2006 Location: Denver Posts: 1,059	I didn't exclude 2002 for "convenience", Heltonfan. I simply am not somewhere that allows me to look at the data for the 140 games that Hurdle managed in 2002. So I only see that the team played 3 games better than Pythagorus in 2002, and can't see that it was 5 games better during Hurdle's reign, but I'll take your word for it. Assuming that's right, the following is the Hurdle record vs. Pythag.... 2002: +5 2003: -4 2004: -5 2005: -3 2006: -5 See any pattern there? If anything, I think 2006 UNDER-reports the number of wins that Hurdle cost his team, and I think you agreed that he was very costly last season. If you have the same inept and incompetent field management, how can you NOT incorporate that into an overall projection of their record? Strictly on a compilation of the numbers, your 79 wins is something I cannot dispute. But, based on Pythagorean AND empirical evidence of Hurdle's self-destructive managerial tendencies, I can and do discount such a projection by a handful of games. It may be subjective to give a haircut to your 79 win projection, but I believe it's proper.

11-23-2006, 12:48 AM	#9 (permalink)
wolf213 Member Join Date: Nov 2006 Location: Thornton Co Posts: 194	I actually think 79 wins is right on, I don't like Hurdle and do think he cost us a ton of games. I would love to see him manage in the national spotlight like the playoffs so the whole world can see how bad he is. I predict drop offs for Atkins and Holliday, but not by much, and they also can very easily stay the same. I also expect improvement from Hawpe and Baker, and if Tulo is even league average, then the SS position would be amazingly improved.

11-23-2006, 09:36 AM	#10 (permalink)
indianadrew Veteran Member Join Date: Nov 2006 Location: indianapolis Posts: 570	isn't it a manrule that we can claim it was a rebuilding season? we still had a better season than 2005 though, right? i think that the most disapointing thing was that we were in alot of the games we lost. it does not seem to me that it is going to take too much to improve from last year. GO HORSE!

11-23-2006, 11:35 PM	#12 (permalink)
rockieprogress Member Join Date: Nov 2006 Location: aurora Posts: 183	If there is any improvement, and I tend to agree that the 3 areas that HF thinks will improve are the most likely, how can they not improve with the poor showing with RISP? Just avg in that area, and they should be able to gain the 3 games. With this off-season playing out as it has, next year is going to be very interesting.

11-28-2006, 12:46 AM	#13 (permalink)
wolf213 Member Join Date: Nov 2006 Location: Thornton Co Posts: 194	Here is my 2 cents. 1. D-Backs 2. Padres 3. Dodgers 4. Rockies 5. Giants I think we can all agree that the Giants arn't going to be very good next year after loosing their Ace, and 2 of their sluggers. D-Backs are looking good in my opinion, they have smart GM who made some nice trades, and they have some good depth in the system Padres are standing pat so far, but I expect them to get a pitcher and maybe another bat. I have learned from season's past never to discount them Dodgers are making dumb move after dumb move. Pierre takes of for Drew? Wolf for 8 Million? Nomar? They have plenty of depth but I'm crossing my fingers they do something really stupid and give up a ton for Manny. Rockies would be ranked higher however Hurdle is still the manager of this team. Even though I have them in 4th I think it will be a very competitive 4th, maybe with a .500 recored.