2007 Season Predictions

[Heltonfan]

Originally Posted by Roxpert

We don't have nearly a league-worst bullpen, but Coors will play more like the "old Coors" in 2007 (I expect around 12.0 to 12.5 runs per game, offsetting the outlier result from 2006).

That means that you're expecting the park-adjusted league-average ERA to be around 5.20 or so (6.00 at home, 4.40 on the road = 5.20 overall). In which case, you've got Cook being one of the top 5 pitchers in the game, Hirsh having a phenomenal rookie year, and Lopez being a rock-solid #2 starter.

And if the bullpen ERA is 5.80 or so, even with Coors playing at 12.5 RPG, it will be the worst bullpen in the league, or awfully close to it. Replacement level for a reliever is only around half a run above the league average ERA.

Your predictions seem to be a combination of rampant optimism and rampant pessimism, with no middle ground whatsoever.

[hiaspire]

Originally Posted by Heltonfan

I'm not saying that your scenario is impossible, but I don't see how we could deem it the most likely outcome (which is what a prediction is supposed to be, after all).

Not necessarily. That's just the safest answer. People make predictions all the time that aren't the most likely. Otherwise everyone would have the same answer always. Everyone sees upcoming events a little differently and has their own slant on what they expect to see. Like my baserunners, I'll take aggressive predictions over the cautious alternative.

[Roxpert]

Originally Posted by hiaspire

Not necessarily. That's just the safest answer. People make predictions all the time that aren't the most likely.

I agree with HiAspire here. When making a prediction, I'm not necessarily trying to make a "point estimate" as I might when forecasting earnings for stocks I own. No real need to "thread the needle" with a W/L prediction in the same manner one does with player projections.

Also, keep in mind that my individual player projections all assume current good level of health. We can't forecast injuries, but they do occur. (Well, I did forecast an injury for Holliday, but that's the ONLY case where I did so). Overall, I expect the 2007 Rockies to have considerably more injury problems than the 2006 Rox, but didn't want to cut Cook's, Francis's, or Lopez's innings to account for such risk.

Regarding W/L actuals vs. Pythag., here is the history since Coors opened:

1995 - Plus 6 (was +5, extrapolated to a 162 game season)
1996 - Plus 2
1997 - Plus 1
1998 - Minus 1
1999 - Even
2000 - Minus 5
2001 - Minus 9
2002 - Plus 3
2003 - Minus 4
2004 - Minus 5
2005 - Minus 3
2006 - Minus 5

The "bomber" teams were the only real "Plus" teams we had. The transition to Buddy Bell was about even (Baylor's last year and Leyland's only year). Buddy Bell was a "Minus 7" manager in his 2+ years. Clint Hurdle's four full seasons (where he was in charge of spring training) has averaged Minus 4.3, and we've averaged a Minus 4 since the Dan O'Dowd era began (2000), covering 7 seasons.

I know luck has a lot to do with variance vs. Pythag., but I also believe managers can influence the result by losing more than their fair share of close games due to poor usage of the bullpen, bunting strategies, etc.

The pattern with Clint is pretty strong. That's why I look for another Minus 5 year in 2007.....he hasn't improved as a manager, and will "run us" into a few more close losses this year.

Edit: I suppose, therefore, that my W/L prediction builds in more things to "go wrong" than my individual player projections which largely look for good health. Truth be told, my original post had the Rox pegged for 75 wins, but I changed it to 73 wins in my first edit. Again, I may revise further once I get more familiarity with the team with the DMB projection disk.

[hiaspire]

If there are teams in baseball who only project the most likely things and make decisions based upon that, they are playing it way too safe and going to miss out on players that other teams take chances on because a skilled professional sees something in the talent that others don't see yet in just running the numbers.

One way is reactionary to the known past only, but finding talent that makes good on future predictions also requires a healthy dose of proactive measures beyond just the most likely utilizing the past exclusively.

The overall math probably treats the safest/likeliest answers well overall when you consider a massive sample, but every off-season there are particular highly-discussed value "predictions" which fall far short of their mark because what averages out sometimes is the most unlikely thing imaginable. Such as Thome and other examples. The calculated prediction might make logical sense in term of weighing different possible outcomes and past extremes, but there is almost no way in the world that a middle-of-the-road prediction/estimation becomes true.

In reality usually teams have to pick one side of the fense instead of going by the average calculation. Some predict more trouble coming, and other teams predicted he'll bounce back to himself once healthy again as he was becoming. Usually you have to be on either side in that less safe ground in order to make acquisitions happen. Teams have have higher-than-average opinions of players versus the "likely" normal levels are the teams that get those players. Same is true on the other side in terms of players a team will let go because he has more interest elsewhere than they have in keeping him.

The likely number sets a good bar generally, but at some point you have to go either above or below because someone else will and you'll be left with the leftovers that nobody has any strong opinions about other than their mediocre likely expectations.

As far as our team, I am way up over the bar on Taveras and Hawpe probably still, along with most highly criticized veteran role guys who I believe will be very beneficial in multiple ways. I am under the bar in terms of our pitchers and Matt Holliday. If they really are messing with the environment that alters these shifting pitching numbers, then I can't see how using those stats alone to make good judgements makes any sense. Just like Gary Mathews Jr and his medicine messes up a numbers-only estimation. Sometimes you have to put the numbers aside and allow for human skepticism with your own eyes in terms of a player's unknown future. In such instances, that certainly wouldn't be the most likely numerically, but certainly serves you better than completely trusting a highly questionable likely prediction calculation.

[BigRapidsJackass]

Our collection of bullpen talent certainly leaves quite a bit to be desired, but it's far from being the league's worst.

I actually quite like the bullpen talent this year.

Fuentes (good) and Hawkins (not so good) are locks.

If Hurdle and Apodaca don't grossly mismanage their talent, Affeldt should be a fine situational lefty (his overall numbers probably won't impress since he should -- and will -- get his fair share of mop-up work; I just don't want him facing righties in critical game situations) and Tom Martin should be an adequate guy for low-leverage situations.

That leaves 3 spots to be filled by the following:

The Reclamation Projects: Graves/Herges/Veres -- I hope none of them make the club. Veres would be a nice story, but that's about it.

The Young Power Arms: Morillo/Jimenez -- I think Jimenez would be a great guy to break in as a reliever. Morillo? The next Ambiorix Burgos. Who knows when? Never?

The Young Power Arms With A Bit of Experience Under Their Belts -- Corpas, Ramirez -- Corpas is less talented than Bautista, but miles ahead in knowing how to pitch. I think he'll be quite good. As for Ramirez: I was seeing a lot of hanging sliders after June last year. And I think I'll be seeing lots more this year. Still, he obviously has the talent if he can master control of that slider.

The Running-Out-of-Chances Power Arm: Bautista. Getting out of KC must be good for any pitcher, right? I think both he and Affeldt may adjust nicely to narrower bullpen roles. He is still a super talent, but this is his last chance ... at least in a Rockies uniform.

The Quirky Delivery Guy Who Won't Go Away: Speier. I think it'll take him a bit to get back arm strength, but if he avoids further arm trouble I think he'll make a nice contribution this year. Eats up righties and appears (from a very small sample) to have major problems against lefties, so he's probably going to be a situational guy, and it's hard to carry more than one of those (Affeldt is a lock) in any pen, but I think his opportunity will come.

The Converted Starters: Kim/Fogg/Buchholz/Lawrence -- Lawrence I think stays a starter. Buchholz may be better suited to relief; he has that reliever's nice 1-2 (plus fastball, plus-plus curve) punch. Fogg relieved in college and could be a very overpriced middle reliever. Kim? Let's face it, he's nuts. He'd still be best in the pen, particularly in a set-up role where a manager could use him in the optimal situations. But he'd probably have a fit. Still, that's his destiny. Accept your fate, BK.

So there's lots of choices. And unlike years past, most of them are pretty appealing. Most of them probably won't work out. Hey, I thought Scott Dohmann would be the surprise of last year. (He was, but not quite in the way I expected.)

The bullpen battle will (as usual I guess) be the most interesting thing going on in Tucson. Let's hope Apodaca and Hurdle can make some wise decisions.

[Roxpert]

RotoRox, you're being a Jackass (sorry, couldn't resist).

You write "I actually quite like the bullpen talent this year", and then go on to write a critique of what sounds like the biggest group of IFs and misfits, and has-beens that I've ever heard of. Just because there are a lot of "choices", as you said, doesn't mean there are that many QUALITY choices.

If I had to rank the candidates for the bullpen on pure skills, from 1 to 8, here's how I'd rank them:

Fuentes - 7 (reliable, close to elite, but not quite an 8.)
Bucholz - 6 (relatively good stuff that should be suited to 'pen work)
Jimenez - 5 (great stuff, too wild, not ready)
Morillo - 5 (see Jimenez)
Affeldt - 5 (ex-starter with million dollar arm and 50 cent head)
Hawkins - 4 (ex-closer past his prime who gets a 4 on experience only)
Corpas - 4 (remember David "Diesel" Lee?)
Ramirez - 4 (he was the righty "Javy Lopez" in '06. Great start, weak end)
Bautista - 4 (poor physical history and hasn't learned to harness his stuff)
Speier - 4 (good at AA in '04, never much of a prospect. Coming back.)
Kim - 4 (won't relieve for us)
Fogg - 4 (will either exclusively start with us or someone else.)
Lawrence - 4 (will only come up if he can start. Not a bullpen candidate.)
Herges - 3 (not much left in that tank.)
Veres - 1 (washed up trash who is damaged goods)
DeJean - 1 (ditto)
Graves - 1 (ditto)

There may be someone else come out of the woodwork (Keppel?), but this is a singularly unimpressive group. When you think about it, a LOT has to go right for this bunch to register a collective ERA UNDER 5.50 whether the '06 humidor is working well or not!

[Heltonfan]

Roxpert, I think your rankings are right on the money (I'd quibble that you're giving Buchholz too much credit and DeJean too little, but in the grand scheme of things, that's pretty irrelevant).

But I think you're guilty of the same mistake you're chiding RotoRox/Jackass (that name really is hard to get used to...) for; the rankings don't support your conclusion. If we have a 7, a couple 5s, a bunch of 4s, and a couple guys in the 2-3 range, that's not such a bad bullpen. Unless you're intending league average to be a 6 or something, in which case the 1-8 scale doesn't make much sense.

[BigRapidsJackass]

That's right, Heltonfan.

There's some good arms out there (yeah, and some total crap, too), and overall there's the potential to put together at least an average pen.

Whether it's a solid pen or not will have a lot to do with how Hurdle and Apodaca manage things ... in who they use in the bullpen vs. starting (Jimenez, Buchholz, Bautista, Affeldt), in who they keep on the 25-man roster, in how they move pitchers between the Sky Sox and the big club (particularly guys like Lawrence and Speier, coming off injury), and in how the starting pitching develops (again, this may free up Buchholz for relief duty).

But the talent is there. And the Rockies are basically in the same bullpen situation as, oh, about 25 other clubs -- they have some proven arms, some good young arms, some old arms who may have a little bit left, and some total garbage.

Kind of like the Red Sox but with a better closer candidate (now that Papelbon is starting) I'd say.

[Roxpert]

Originally Posted by Heltonfan

But I think you're guilty of the same mistake you're chiding RotoRox/Jackass (that name really is hard to get used to...) for; the rankings don't support your conclusion. If we have a 7, a couple 5s, a bunch of 4s, and a couple guys in the 2-3 range, that's not such a bad bullpen. Unless you're intending league average to be a 6 or something, in which case the 1-8 scale doesn't make much sense.

My 1 - 8 rankings were in terms of overall skills/talent (i.e. stuff). I would translate them as follows (for the Rockies' context):

An "8" would be capable of "Gabe White 2000" results. ERA low to mid 2's.
A "7" is a reliable pitcher who can blow up on occasion. ERA High 2's to 4.00.
A "6" is a streaky pitcher who can look like an 8 or 4. ERA mid 3's to mid 4's.
A "5" has above average stuff but inconsistency. ERA low 4's the low 5's.
A "4" has purely average stuff, lots of baserunners. ERA high 4's to high 5's.
A "3" is a flawed MLB reliever without job security. ERA mid 5's to mid 6's.
A "2" is on the bubble between MLB and AAA. A "filler" - ERA 6+
A "1" is a has-been or never-was, due to injury/age. ERA 7+

Assuming our managers/coaches know just who the #4's are beforehand, and avoid giving any of the 1's through 3's innings, I agree that the bullpen ERA could be low to mid 5's. I don't think our field braintrust is that smart, and will try to force Herges, possibly Veres, or other drek into too many games.......and high leverage situations (Herges alone could cost us 3+ wins by having putrid 7th innings in close games). Thus, I believe that the bullpen will come in with a poor ERA this year. Fuentes won't be quite as good as before (ERA about 4) and this will prompt the trade, along with his contract.

The other problem is that the above-average talent I have ranked "5" (other than Affeldt) is not quite ready yet to contribute to the bullpen this year. Maybe they will get innings in July/August, but I wouldn't expect much from them despite their relatively lofty rankings.

[Heltonfan]

(Herges alone could cost us 3+ wins by having putrid 7th innings in close games).

No, he couldn't.

Suppose Herges' ERA is a full run worse than league average. That's well below replacement level, and significantly worse than I have him projected at... certainly a pessimistic assumption.

Now, let's say that Herges pitches 81 innings, at a Leverage Index of 2.00 (this is insanely high... closer territory, basically). At that rate, he'll cost us 9 runs (relative to average) over the course of the season. Multiplied by 2 for the leverage, that's the equivalent of 18 runs, or nearly two wins.

And again, that's based on a pessimistic assessment of his talent level, a pessimistic assessment of Hurdle's ability to use him correctly (even Clint isn't going to run a guy with a 6+ ERA out there in the most critical situations all year long... yes, he did it with Chacon, but that was a special case), and an utterly ridiculous assumption about the leverage of situations faced by a 7th-inning guy.

People dramatically overstate the importance of bullpen usage. Is it significant? Sure. Can one reliever cost the team three games pitching in the seventh inning? Not a chance.

[hiaspire]

Originally Posted by Heltonfan

People dramatically overstate the importance of bullpen usage. Is it significant? Sure. Can one reliever cost the team three games pitching in the seventh inning? Not a chance.

That better be the theoretical run stuff again or it is way off base IMO. Absolutely a reliever can cost his team ACTUAL games all the time. I know it must be that theoretical stuff because actual game losses can't always be pinned on just one individual, of course, but sometimes it can be pretty darn close when the team is ahead late and a reliever self-destructs on his own on the mound to completely turn the game around and lead to the loss from victory's grasp.

That is costing a team a win MUCH more than whatever calculation is saying it because it is real. And absolutely a reliever could do that to his team many times. Just ask Chacon or Mesa or countless other blowups-waiting-to-happen that have come through the pen.

[Roxpert]

Interesting analysis, Heltonfan, and you may be right. However, I also think (though I failed to write it) that Herges will also see the 8th inning in quite a few outings. He may mainly be a 7th inning guy, but getting 30 or 40 appearances in the 8th is entirely possible (depending on how Hawkins is doing).

[RMF]

I've got some more predictions.

Clint Barmes will make huge advances offensively this season. He'll be good enough to fend off Tulowitzki until the end of 07. At which time he'll be moved to either 2B or CF. He's going to set career highs in every offensive catefory.

The surprise of the year, he'll hit 20+ Hr's. .279/.337/.456

Todd Helton will have a very good year hitting 30+ Hr's. He'll take an occasional day off and play things smarter this season. His rate statistics will be awesome. .330/.427/.550

Brad Hawpe will squelch all these platoon rumors. Even though he won't hit for average against LHP he'll still crush the ball and play good defense. Baker will get occasional starts in the corners and at 1B.

[Heltonfan]

If Herges is used half in the 7th inning and half in the eighth, he'd probably have a Leverage Index of 1.5 or so. In which case, we're still only at ~1.4 wins below average. You can't get close to 3 wins below average without making some absolutely preposterous assumptions. That's just a mathematical fact.

Since the Pythag topic has come up again, I decided to take a look at Hurdle's record in one-run games. Bill James did a study a while back in which he found that a team's record in one-run games can be estimated by the Pythag formula, using .865 as the exponent instead of 2. With that as a guide, here's how we've performed in one-run games during the Hurdle era, measured by wins above/below the modified Pythag:

2002: 0.6
2003: 2.1
2004: -3.1
2005: -2.1
2006: -3.0

Total: -5.5 wins, over 5 years. Average of one win a year. I'd be comfortable ascribing that difference to Hurdle rather than to luck. But to blame Hurdle for the entirety of the Pythag underperformance is a real stretch. We have to keep in mind that Pythag differential is just as much a function of a team's performance in blowouts (which say nothing about the manager) as it is of a team's performance in close games.

[wolf213]

Roxpert, I think you are undervaluing Corpas by a ton. He is young still, but everything from the Minors and his limited work in the major scream future setup guy or even closer. Why so down on him?