Daily Fantasy Strategy – 7/22/13 – For Draftstreet
I love the commenters on this site. I write elsewhere, and the audience is more interactive here than anywhere else I’ve seen. And it’s not just that they’re involved; it’s that their they’re smart. I’ve gotten several ideas for posts from commenters, and they frequently point out things that I miss. But I think they may have one thing wrong. Or at least there may be something upon which they’re placing too much weight.
When picking pitchers, the first thing I look at is the quality of the pitcher’s opponent against pitchers of the same handedness. Well, I take that back. The first thing I look at is whether the pitcher is any good or not. But I don’t usually have to look that up. After those initial hurdles are cleared, I look at conditions. I’m looking for cooler temperatures and winds blowing in (with a few exceptions). All of these factors generally go toward run prevention.
But commenters often point out a pitcher’s strikeout ability and/or the strikeout percentage of the opposing team. They note the importance of strikeouts on Draftstreet. I myself used to look at the opposing team’s K% as one of the more important factors, but I seemed to have more success when I placed most of my attention on run prevention. But was I right to favor run prevention over strikeout potential? Or are the commenters right? Is strikeout percentage the most important factor?
To answer that question I simply tested the relationship between average points per game for pitchers in Draftstreet scoring to both ERA and strikeouts. Below are two charts showing the two relationships with the relationship with ERA on the left and the relationship with strikeouts on the right.
Any time I start messing around with correlations and scatter plots and R2 numbers, there’s a chance I’ve done something horribly wrong. I’m trying to catch up on the whole statistics thing because I paid very little attention in stats class in college. But assuming I’ve done this correctly, it seems like earned run prevention has a much stronger correlation with Draftstreet points that strikeouts do. But that’s assuming I’ve done this correctly and assuming I’ve not missed something obvious. But if I have missed something, I’m sure the commenters will set me straight.
The Daily Five
Left-handed Colorado hitters, varying prices – Colorado is good for hitters. Hot weather is good for hitters. Right-hander Tom Koehler and his 4.70 ERA is good for hitters. Here is a list of the Rockie hitters that will be hitting from the left side (if they’re in the lineup) and who will be facing Tom Koehler in Colorado with a game time temperature in the high 80’s: Carlos Gonzalez (OF, $10,918), Todd Helton (1B, $7,215), Dexter Fowler (OF, $9,074), Jonathan Herrera (SS, $5,397), and Charlie Blackmon (OF, $6,674).
Right-handed Miami hitters, varying prices – The conditions are the same for the Marlins in Colorado today. The only difference is that the Rockies have a left-hander on the hill, Drew Pomeranz. Here are the right-handed hitters to consider using for Miami today: Adeiny Hechavarria (SS, $4,671), Marcell Ozuna (OF, $6,675), Giancarlo Stanton (OF, $8,074), and Justin Ruggiano (OF, $6,312). Stanton and Ruggiano seem like good values at those prices with their power. And Hech is so cheap that it’s going to be hard to pass on him.
Max Scherzer, $18,806 – Another thing the commenters don’t like is spending this much money on a pitcher. And it’s a valid concern. If you can find good pitchers with good matchups with a lower price tag, you should. But the options aren’t great today. All of the best matchup teams are either facing a bad pitcher or are facing a pitcher of the handedness that the team happens to be better against. And a lot of the better pitchers today have bad matchups. So I’m paying for the safety of Scherzer against a bad White Sox lineup.
Matt Garza, $15,741 – The roof in Arizona should be closed so as to avoid the 90 plus degree heat. And cooler temperatures are always good for run prevention. It’s also a good matchup for Garza against the D’Backs who have the 6th worst wRC+ vs. RHP.
Julio Teheran, $12,366 – Reasonable price? Check. Good conditions? Check. Temperatures in the high 70’s and the wind projected to be blowing in from right. Good matchup? Check. The Mets have the 8th worst wRC+ against RHP. They also have the 3rd highest K% vs. RHP, which I’m sure the commenters love.
One Day FREE Fantasy Contest – $300 in cash prizes
Daily Fantasy is easy — you pick a lineup that’s good for one night only, and then you’re eligible to win cash. FanGraphs readers are hereby invited to enter a FREE one-day fantasy league with $300 in prizes.
This free contest will be Pick ’em style drafting. The way Pick ‘Em leagues work is you have eight tiers of players and each tier will have players to choose from. You select one player from each tier. Nothing to lose and it takes five minutes to build a team. You can adjust your roster up until the contest starts on this Friday, at which time your rosters will lock and the Live Scoreboard will be available.
This post, covering one of the leading sites for daily fantasy is sponsored and made possible by the generous support of Draftstreet. FanGraphs will maintain complete editorial control of the postings, and brings you these posts in our continued desire to provide the best analytical information on the latest in baseball.
You can find more of Brett's work on TheFantasyFix.com or follow him on Twitter @TheRealTAL.
It’s that they’re* smart.
Lol – glad I’m not the only one that caught that. Must be our collective inherent ‘smartness’ on display..
Lol and clearly I am not.
while we’re ruthlessly nitpicking, you really should label the axes on your graphs, like all the time
Dumb question I know, but how do you do that?
In Excel, when you’ve selected your chart, you should get a “Chart Tools” tab to appear. Then under the “Layout” tab there is an “Axis Titles” button.
I have to preface my comments by saying I’m not an expert in the statistics myself. I think you’ve asked a real interesting question. But something jumps out at me right away that might help you in the future.
To me, these graphs would answer the question “Is season ERA a better indicator of per game score or is season total strikeouts a better indicator?”. So I don’t think that these graphs answer the original question you raised.
Your original question mentioned K%, but the graph looks like it uses strikeout count, not a rate.
I think you would also have to do run this on an individual game basis to get the true answer to your question. Not on season ERA as a whole. Run the scatter plot of ERA for each individual game vs. the score for that individual game. But even that still doesn’t isolate the effect of the weather factors you mention.
There’s got to be a study out there somewhere that tests the effect of wind speed and temperature on run production. It won’t be directly to draft street game score, obviously, but would give you some idea the effect. Your current analysis is on ERA as a whole, not on the marginal effect on ERA that weather factors have on it. You somehow would need a way to isolate only the effect of the weather.
Not familiar with the draft street scoring system off hand, but if you can find a study of weather effect on run scoring output, you could just plug that into the scoring system to understand the true impact on game score.
Not sure who originally posted it, but one of the other daily guys posted a weather study awhile back. And Blake Murphy has been looking at a lot of wind speeds lately. But I think you’re right on a lot of this stuff. At some point I’ll go back and redo the study.
Always read the Daily even though I don’t play daily because I like the insights that can be used in my league. One thing I have NEVER thought of is the Arizona roof closure. I would LOVE to see a study about the effects of that and other roof closures around the league. Arizona is typically known as a hitters park. I wonder if this changes when the roof is closed! Please do the study or beg someone to do it. This would be a HUGE hidden advantage to us Rotograph readers.
In a nutshell, I focus on both opposing team’s offensive production vs a handedness of a pitcher, and an opposing teams strikeout rate. I never really look at one or the other.
I think the study needs to include not just ERA, but hits and walks since those are negative points in DS.
I think Mondays provide us with an opportunity to pick pitchers who are at home that are facing teams that just traveled. IE, Yu Darvish vs the Yankees. Not sure I’ll pay up for Darvish today as I haven’t gotten into my research, but I do like picking on teams fresh off a road trip when I’ve got a team on a homestand.
I wouldn’t start Garza today ;-). Also wouldn’t start Teheran. Looks like rain. For a bit lower price, I’m starting Darvish. Better win candidate since he doesn’t face Sale. Sometimes that 1.5 pts does come in handy. Yankees offense blows and they had to hop on a plane at like 2am or whatever it was last night after bitter extra innings defeat at Fenway. And I say this as a Yanks fan, start Darvish.
Love these columns.
Two questions, one related to this column. You talk a great deal, especially in this post about ERA relative to Draft Street success. How much do you factor in other numbers like xFIP or SIERA? This seems to be an area to identify additional bargains, especially if they are out/under performing their peripherals. Joe Blanton at a devilish $6,660 against a below-average wRC+ Twins seems to be the kind of bargain to look for.
Another question unrelated to this that maybe you can touch on at a later date regards timing/roster construction. When do you usually pick your lineups? After manager cards are in and DS is reporting who is essentially starting or sitting? If you choose early, what if a player you’ve chosen ends up not playing? Do you try and replace him with a similarly priced player who may be a significantly worse value or do you rearrange things across the board to make value fit into the salary cap? If you’ve already touched on this, a link would be great. Thanks for the articles.
When I was talking about first considering whether a pitcher is any good or not, xFIP and SIERA are a big part of that determination.
I have to pick my lineups the night before in order to write the column. But I usually check back around an hour before the first games to check starting lineups and more updated weather conditions. If someone isn’t playing or it looks like the game is a real risk to be rained out, I’ll first just try to replace the players that are not or might not be playing with similarly priced players. But sometimes it’s just not doable, so sometimes I have to essentially blow up the team and start over.
Twins have hit almost .300 v. Blanton with a HR every 15 at bats.