02-03-2014, 03:10 PM #141
- Liked 214 Times in 147 Posts
I think the offense was hard to fix this year....there were either bad players, or elite players available, or 1B/3B types. Apparently we don't ever need another OF here, as the minors will fix that for us for sure......so there weren't a lot of good offensive guys to sign. And, that's how it is going to be, imo. Medium guys will be signed to extensions. Only expensive, cheap, or old/injured guys will be in FA in the future (offensively).Lighten up Francis....
02-03-2014, 05:51 PM #142
And ....it also turns out that my own net .825 OPS projection from 1B was very optimistic. I would argue that their projections from .745 OPS/DH and .697 OPS/C are similarly even more optimistic and wildly more conditionally-based....RE-
unless Pinto moves into the full-time role sooner rather than later....AND....hits like he did last September. This is a very optimistic forecast, that to this point doesn't fully reflect the very potential reality of the team starting out the season with Pinto in AAA and Suzuki as the starter and either/or Fryer/Hermann as the backup(s).... I would argue that my number was more realistic.
and Jason Kubel gets 400+ PAs at DH and somehow semi-miraculously beats his average projection from 3 services (Steamer/Oliver/ZIPS) of .706 OPS-----the only way this .745 number is realistic is if Willingham successfully rebounds AND gets significant PAs at DH, AND that Mauer is factored in for far more DH ABs than I project. The DH net number I projected- being closer to .700 OPS, again seems more realistic.
Last edited by jokin; 02-03-2014 at 06:08 PM.
02-04-2014, 08:47 AM #143
You're more than welcome to disagree with the playing time projections that Fangraphs is using. However, I don't think you have any basis at all to claim that ZIPS and Steamer are anything other than completely neutral.
I'm not sure how ZIPS/Steamer having Pinto at a .723 is hitting "like he did last September", but let's not let hyperbole and selective data get in the way of making a point... certainly no shortage of that elsewhere.
Let's compare, 1B/DH/C for 2014:
Fangraphs - .751
Me - .737
You - .725 and that's "very generous" with copious usage of "optimistic"
02-06-2014, 10:28 AM #144
- Liked 22 Times in 15 Posts
If there are injuries to any of the FA pitchers then it will be a 90 loss season since the offense wasn't improved perhaps another 96 loss season.
02-06-2014, 08:08 PM #145
To wit: Oliver and ZIPS have Pinto somehow playing 143 games. At this point, Steamer probably has taken the more realistic assumption with 81 games as their guess. Average the 3 together and you come up with a .716 OPS for Pinto.
I'm going with Steamer and the indication of the Twins intent, plus Pinto's injury and Winter ball shutdown (Ryan has announced that Suzuki is the Twins opening day catcher and Pinto is indicated to be headed to AAA, at this point, anyway)----
while you're going with the other 2 services out of at least a half a dozen more that disagree with both you and them.
Suzuki, among the 3 services I used, averages to a projected OPS of .656, with Steamer giving Suzuki the extreme outlier, higher projection, of .677 playing half-time. Every other projection out there, except Steamer, has Suzuki OPSing between .623 and .645.
What was Suzuki's combined OPS for 2012 and 2013, you ask? .614!
Again, who is being optimistic here, who is being hyperbolic and who is being realistic?
If you assume that Pinto stays down until the All Star break, that would mean that Fryer/Hermann play in 20 games for around 80 PAs at a .611 OPS.
Assuming an 80/20 split of the PAs to the break, the combined OPS at Catcher would be .645. With makeup games added there will be about 70-72 games left after the All Star game, so let's say that Suzuki is then traded- that leaves 60 games for Pinto? Or...let's assume, very generously and optimstically and probably unrealistically, that Pinto gets called up even sooner than the All Star game, say in mid- June, Fryer is DFAd instead, and Suzuki is basically relegated to one or two days a week duty for the rest of the season- and Pinto STILL ends up playing 60 games at Catcher. According to the projections, that would mean that the combined OPS at the position would be around ~.655- 670 OPS.
This generous and optimistic forecast at the Catcher position would mean that, in order to hit your own combined OPS number at C/1B/DH, that Kubel has to come through above the lofty projections despite his very questionable status, and/or that Willingham also:
comes back strong from his injury-plagued-year,
gets significant DH PAs,
AND is not traded away in June or July.
A lot of things have to hit just right for your scenario to come to fruition, I just don't see it. And the combined OPS number comes in right around mine if we hit these rosy forecasts and only end up getting .811 OPS combined from the 1B postion as is projected in your own scenario.
Last edited by ashburyjohn; 02-06-2014 at 11:21 PM.
02-07-2014, 07:33 AM #146
- Liked 179 Times in 106 Posts
Then, to project standings, Fangraphs or Clay Davenport or whoever will make subjective estimates on playing time. Davenport has him at 176 PA, and Suzuki at 358:
Fangraphs has Pinto at 320 PA, Suzuki 288:
I might guess the former split is more likely, but referencing the 563 ZiPS number in any critique is completely meaningless.
02-07-2014, 07:51 AM #147
- Liked 33 Times in 25 Posts
So, two "experts" disagree on how many PA's the two catchers get, how really important is that in the whole production or lack thereof in the entire lineup? I would think PA's plus production in traditional hitting spots in the entire batting order is much more important than wondering what two rather pedestrian catchers give to the lineup. If the so-called "big guns" don't produce 90+ losses again are staring us in the face.
02-07-2014, 09:15 AM #148
Win Ratio = 4*(OPS/OPS Allowed) - 3
So, a team that is perfectly average in both OPS for and against has a win ratio of 1 with an expectation of 81 wins. We've been debating 3 positions summed together, so a third of the lineup. If you change the production from those 3 spots by .025 (for example, .725 vs .750), it would affect the team OPS by a third of that. Let's call .725 the average and see what impact it has:
WR = 4*(.733/.725) - 3 = 1.046
Win Ratio can be converted to a more understandable winning percentage easily. W% = WR/(WR+1)
W% = 1.046/(1.046+1) = .511
162 * .511 = 82.8
So, the team that was perfectly average saw their expected wins go up by 1.8 by getting a .750 OPS instead of .725 from a third of their lineup.
Last edited by jay; 02-07-2014 at 09:26 AM. Reason: convert to W%
02-07-2014, 11:11 PM #149
Why not just release the rate stats without all of the further detail if it's so meaningless?
02-07-2014, 11:57 PM #150
- Liked 74 Times in 50 Posts
Thank you for responding.
The referenced formula seems simplistic and has no qualifiers--I'll use it for discussion purposes. The expected 1.8 additional wins generated by this incremental OPS would be well within the error margin of the original estimate of wins postulated in the thread and likely within the error limit of the cited formula itself! Ergo, noise.
02-08-2014, 12:43 AM #151
You're probably right that either result is well within the error range for both scenarios. However, if that's your definition of what makes it noise, there wouldn't be a single move you could make that wouldn't qualify. Even a big free agent signing and the expected gain in wins would be well within the error range of what could actually happen without that FA given all the dynamics of an individual season.
If you played the season a million times, you'd find these teams would get extremely close to averaging the 81 and 82.8 numbers respectively. Although individual seasons are within error range of each other, that expected number doesn't change.
02-08-2014, 03:35 PM #152
- Liked 179 Times in 106 Posts
Basically, depth charts, roster composition, and health reports are not direct inputs into ZiPS. I think this is covered in most overviews of the system.