Detailed Player Analysis - Fantasy Baseball Cafe 2015

## Detailed Player Analysis

Moderator: Baseball Moderators

garf112 wrote:If he was at HIS average for gb/ld/fb? or at the league average? I was under the assumption that a hitter DOES have control over those percentages and tends to hit roughly the same amount each year, with some peaks and valleys added in.

League average. The number isn't meant to represent what we expect from the hitter as far as batting average goes. It's meant to express what batting average you would expect from a player with a .300 BABIP that put up those same k/hr/ab numbers.

The reason it's designed that way is so we can see how well the batter's plays determined at home plate support a good batting average. By expressing it versus the league average you do two things - you standardize the number and you make it easier to determine as well since you don't have to determine the expected BABIP for each hitter. It's not intended to use in projecting batting average...it's more to give a second view of the batting average variability. The (k + hr) / ab% gives us the range of values while the batting average using the league average gb/fb/ld percentages lets us know how good the hitter is at keeping his outs in the field and his hits out of the park.

The one thing I'd consider too for the k + hr / ab% is trying to find different weights for the k's and hr's too...a k is much less important than a hr and needs to be represented as such.
The Loveable Losers
Hall of Fame Hero

Posts: 7290
Joined: 30 Mar 2005
Home Cafe: Baseball
Location: Cubs Win!!! Cubs Win!!!

By the way, the two players that pop right into my head when I think about this project are Adam Dunn and Ichiro Suzuki. Dunn would have a very low expected average but at the same time have a very low variability (since every at bat for him seems to end in a home run or a strikeout). Ichiro on the other hand would have a decent expected average but a VERY high variability - he rarely strikes out or hits a home run.

That variability has been on display with both of them. Dunn always plods along within about .016 points of his career average. Ichiro on the other hand can fluctuate about .035 points either way over the course of a season.
The Loveable Losers
Hall of Fame Hero

Posts: 7290
Joined: 30 Mar 2005
Home Cafe: Baseball
Location: Cubs Win!!! Cubs Win!!!

As a general rule, I've found it real difficult to project MR production year-to-year. Scott Linebrink and Scott Shields are usually money, but most get so few appearances, that often vulture wins are as much about luck as talent. One guy I like is Mike Myers. He's usually just a get-a-lefty-out 1 or 2 batter appearance guy so he doesn't get the innings that most good MR guys get, but he's got great #s versus lefties, and he can rack up a lot of holds when the Yankees are on a roll where they're winning by just a few runs.

Other than the few constantly good MRs, I've found it's usually a wait and see affair because they come and go so much.
johnsamo
Major League Manager

Posts: 1826
Joined: 30 Mar 2006
Home Cafe: Baseball
Location: Los Angeles

johnsamo wrote:As a general rule, I've found it real difficult to project MR production year-to-year. Scott Linebrink and Scott Shields are usually money, but most get so few appearances, that often vulture wins are as much about luck as talent. One guy I like is Mike Myers. He's usually just a get-a-lefty-out 1 or 2 batter appearance guy so he doesn't get the innings that most good MR guys get, but he's got great #s versus lefties, and he can rack up a lot of holds when the Yankees are on a roll where they're winning by just a few runs.

Other than the few constantly good MRs, I've found it's usually a wait and see affair because they come and go so much.

When I projected the wins totals for guys using this it was fairly close (within about 3-5% per year) on the overall total. Obviously individual outliers exist but I usually only had less than 10% of the MR's that were off by 3 or more wins. Most of them were within 1 win of being correct.
The Loveable Losers
Hall of Fame Hero

Posts: 7290
Joined: 30 Mar 2005
Home Cafe: Baseball
Location: Cubs Win!!! Cubs Win!!!

The Loveable Losers wrote:League average. The number isn't meant to represent what we expect from the hitter as far as batting average goes. It's meant to express what batting average you would expect from a player with a .300 BABIP that put up those same k/hr/ab numbers.

The reason it's designed that way is so we can see how well the batter's plays determined at home plate support a good batting average. By expressing it versus the league average you do two things - you standardize the number and you make it easier to determine as well since you don't have to determine the expected BABIP for each hitter. It's not intended to use in projecting batting average...it's more to give a second view of the batting average variability. The (k + hr) / ab% gives us the range of values while the batting average using the league average gb/fb/ld percentages lets us know how good the hitter is at keeping his outs in the field and his hits out of the park.

The one thing I'd consider too for the k + hr / ab% is trying to find different weights for the k's and hr's too...a k is much less important than a hr and needs to be represented as such.

When it comes to batting average Ks and HRs have the same weight. A K is 0-1, and a HR is 1-1. Automatically.

So the idea is to find out how prone to BA variability a player is?
Or just to find out how much better than the league average player he is regarding gb/ld/fb%. This seems like a circular arguement you are making.
Join Pennant Wars today! A simulated fantasy baseball game where you are the General Manager.

http://www.pennantwars.com/index.php?r= ... &user=2163
garf112
General Manager

Posts: 2307
(Past Year: 38)
Joined: 10 May 2005
Home Cafe: Baseball

garf112 wrote:When it comes to batting average Ks and HRs have the same weight. A K is 0-1, and a HR is 1-1. Automatically.

So the idea is to find out how prone to BA variability a player is?
Or just to find out how much better than the league average player he is regarding gb/ld/fb%. This seems like a circular arguement you are making.

On your first point a k and a hr don't automatically have the same effect on batting average. A strikeout is definitely an automatic 0-1...that part is correct. And a home run is a 1-1...also correct. But if a batter does not strike out there are two outcomes - a ball in play and a home run. On the home run outcomes the batter obviously hits 1.000. However a batter does not hit .000 on the balls in play...rather the average hitter hits about .300 on those balls in play. So while striking out gives a hitter an automatic decrease from their regular average to .000 getting a home run does not mean an increase from a .000 average to 1.000. It means that the batters is going from a league average of .300 on balls in play to a 1.000 on the home run.

The difference between a ball in play outcome (.300) and a strikeout (.000) is much different than the difference between a home run outcome (1.000) and a ball in play outcome (.300). Just as each hit is more significant to a player's average than each out, each home run is more significant to a player's average than each strikeout...at least as long as the player hits under .500 for the year.

On the second question the main thrust of the project would be to measure the variability of a player's batting average simply based on luck on balls in play. Finding out how much better or worse than average they are in using their balls not in play to support their average would just be a secondary part of the project.
The Loveable Losers
Hall of Fame Hero

Posts: 7290
Joined: 30 Mar 2005
Home Cafe: Baseball
Location: Cubs Win!!! Cubs Win!!!

pre-draft next year, do you plan on doing this stuff again?.... If so, I'd love to see it. Other than real basic stuff like Ks and walks per innings, I don't have the time to look deeper into MR stats. I generally grab Shield and or Linebrink early and focus more on hold guys over save guys, so your research could help a lot.

My league counts Wins, Ks, Saves, Holds, ERA and WHIP, and my usually strategy is, have dominant high K, low ERA and WHIP SPs (because they throw most of the innings) and on saves and Holds, I try to get one Ace in each category, but partially punt saves after that because there is typically a run on save guys and they get drafted way too early than they should because people panic, so I like to counter strategy and focus on hold guys who'll also get vulture wins and won't drag me down in ERA and whip....,

My pitching strategy is, excel or dominate in Wins, Ks, holds and ERA and WHIP, and stay out of the cellar in saves, since it's so hard to predict saves anyways with injuries and cold streaks, etc,. and since everyone else is focussed on saves more, it's hard to predict where you'll wind up ranking anyway even if you focus on saves in the draft. Plus, there are always save guys who emerge that nobody saw coming during the draft.

At worst, excelling in Holds and tanking saves evens out to a wash and I do great in the other categories, but generally, if you don't get the injury bug, I've found that one ace reliever and a 2nd tier guy along with a pickup here and there can keep you out of the cellar in saves.
johnsamo
Major League Manager

Posts: 1826
Joined: 30 Mar 2006
Home Cafe: Baseball
Location: Los Angeles

johnsamo wrote:pre-draft next year, do you plan on doing this stuff again?.... If so, I'd love to see it. Other than real basic stuff like Ks and walks per innings, I don't have the time to look deeper into MR stats. I generally grab Shield and or Linebrink early and focus more on hold guys over save guys, so your research could help a lot.

My league counts Wins, Ks, Saves, Holds, ERA and WHIP, and my usually strategy is, have dominant high K, low ERA and WHIP SPs (because they throw most of the innings) and on saves and Holds, I try to get one Ace in each category, but partially punt saves after that because there is typically a run on save guys and they get drafted way too early than they should because people panic, so I like to counter strategy and focus on hold guys who'll also get vulture wins and won't drag me down in ERA and whip....,

My pitching strategy is, excel or dominate in Wins, Ks, holds and ERA and WHIP, and stay out of the cellar in saves, since it's so hard to predict saves anyways with injuries and cold streaks, etc,. and since everyone else is focussed on saves more, it's hard to predict where you'll wind up ranking anyway even if you focus on saves in the draft. Plus, there are always save guys who emerge that nobody saw coming during the draft.

At worst, excelling in Holds and tanking saves evens out to a wash and I do great in the other categories, but generally, if you don't get the injury bug, I've found that one ace reliever and a 2nd tier guy along with a pickup here and there can keep you out of the cellar in saves.

I'll definitely be doing some of this stuff next year. My plan now that I have quite a few new tools at my disposal is to take Rotowire's pre-season projections as my starting point. I don't have the experience or inside knowledge to predict playing time and they use Pecota for that. I also don't have a lot of experience with predicting the ebb and flow of guys k/hr/ab or k/bb/hr/ip component numbers based on where they are in the age/experience curve - something that Rotowire does a very good job of handling for me.

Once I have their numbers I'll project each team's expected scoring by taking the projected RC/27 outs for each hitter, dividing it by 9 and totaling it up. It's a bit of a rough approach but it's something I can bang out in an hour or so and should get me in the ballpark. I'll of course toss in DICE and pWhip into my pitcher projections to make sure Rotowire didn't do anything foolish with their ERA/WHIP projections. Then I'll toss in a formula to project decisions based on IP and role. I'll probably override the formula on a few pitchers but I'll try to stick close to the 10.099 number (or whatever it ends up being when I toss 2006's final stats into the analysis as well) for middle relievers. Finally I'll use a simple lookup to find out their team's run support and run the pythagorean expectation to give me that pitcher's expected winning percentage. I'll use those two numbers to project the wins (which along with the decisions I'll leave in fractional form) and then determine the IP / Win from there.

I'm not comfortable making all of this available on the boards here as Rotowire's projections aren't mine to publish. I would be happy though to post my IP / Win numbers based off of those projections once I have them available. It shouldn't take me more than a week or two after Rotowire's projections come out depending on how busy I am at that point.

Edit: One other thing I'll look at is the difference between my win projections and Rotowire's. While I can't post what Rotowire's projections were I can certainly let people know whether their projections for wins match up well with my own.
The Loveable Losers
Hall of Fame Hero

Posts: 7290
Joined: 30 Mar 2005
Home Cafe: Baseball
Location: Cubs Win!!! Cubs Win!!!

First the Projecting Wins and Projecting Whip articles have both been posted to the Cafe's articles section. Thanks to a reader's comment the whip projection formula has been tweaked which will allow for greater accuracy - especially when adjusting for different BABIP.

Second I have the new article up.

Alfonso Soriano - Which One is Real?

Excerpt:
The million dollar question is which Soriano is the real one. I've laid out three scenarios here and given the rationale for each...my recommendation would be to read and consider each one and make your decision based on whichever rationale makes the most sense to you. Whatever you believe though Soriano certainly made his owners VERY happy for the first two months of the year. Hopefully you didn't see those numbers and trade for him after that big day on June 4th expecting more of the same.
The Loveable Losers
Hall of Fame Hero

Posts: 7290
Joined: 30 Mar 2005
Home Cafe: Baseball
Location: Cubs Win!!! Cubs Win!!!

Bravo! I don't own Soriano, and I wasn't going to trade for him, but that was such a good article that it just kept me reading.
Join Pennant Wars today! A simulated fantasy baseball game where you are the General Manager.

http://www.pennantwars.com/index.php?r= ... &user=2163
garf112
General Manager

Posts: 2307
(Past Year: 38)
Joined: 10 May 2005
Home Cafe: Baseball

PreviousNext