MLB FORUM

Comments

  • CruzinToVictory

    RG Product Coordinator

    • Blogger of the Month

    Hey Grinders,

    I’m extremely excited to announce the release of the Beta version of my personal MLB model! I call it the Complete Rating UtiliZation Model, or CRUZ (I know, I’m so clever!). I’ve been hard at work tweaking and honing this thing all season, and while it’s an ever evolving process, I feel like it’s finally ready for a public beta version.

    Note/Disclaimer: Please remember that this is, in fact, in the early beta stage. I’m releasing it in it’s current form as a free tool to acquire feedback before next MLB season. With that in mind, enjoy!

    You can find these models in the Daily Research console, or you can go directly to the DraftKings version or the FanDuel version.

    There are a ton of factors that go into this model, but here are some of the individual stats and factors that go into each rating:

    • Vegas odds
    • Batter splits (L/R, H/A)
    • Batter stats for each split, including: K%, BB%, ISO, and PFwOBA (this is a stat that takes Park Factors into account when calculating wOBA)
    • Pitcher stats for each split, including: K%, BB%, wOBA allowed, SIERA, Batted Ball profiles (GB%/FB%)
    • Bullpen matchup (ERA, K%, SIERA)
    • Position in the batting order
    • Surrounding wOBA – This stat is unique to my model. It pulls the actual lineup (or projected lineup, if it isn’t confirmed yet), and averages the wOBA of the batter before and after the hitter in question to give an idea of how much support he has around him.

    I also wanted to include a bit more emphasis on the bullpen, as I believe it is often extremely underrated when looking at MLB DFS. So, I’ve partnered up with RG’s own Paul Sporer, who’s resume includes our “Sporer Report” MLB series and the Pitcher Breakdown last season. Paul has an unmatched knowledge of MLB pitching rotations, and every morning, he manually adjust each team’s bullpen rating based on many factors, such as which closers will be available that night.

    Model Types

    At the launch of this beta, we have three different model “types” that weigh each category differently depending on the goal of that model. Long term we will be automating this process so that you can adjust the weights as you see fit, but for now we have three versions:

    • A “Balanced” model (this is the one I personally use the most)
    • A “Cash” model that weighs things like Vegas Odds and batting order more heavily
    • A “GPP” model that weighs high upside stats like ISO more heavily

    General Notes

    I’ve tried to include just about everything you might want all in one place while using a model such as this. If you simply want a list of players and ratings, that’s right there for you. But if you want to dig a little deeper, I’ve shown a metric ton of stats for every player based on their matchup, so be sure to check out the Legend if you’re not sure what a particular column means.

    Many columns are also color coded for easy comparison between columns. Blue is the best matchup, followed by dark green, light green, yellow, orange, and red.

    The included players are based on confirmed lineups (indicated by the green check mark), but if there is not a confirmed lineup for that team, it will use the RotoGrinders projected lineups from the MLB Lineups page.

    Be aware that I’ve weighted down players with extremely low sample sizes, but there are instances where a low sample size might inflate or deflate a player’s rating. I’ve shown the number of plate appearances so you can take this into account quickly.

    These models are updated periodically as lineups come in, so be sure to check the “Last Updated” note at the top of the page.

    Feedback

    As we’re nearing the end of the MLB season, I’ve decided to release this in it’s current beta form to get feedback from you all so that I can make even more improvements for next season. Take a look, let me know what you think, and leave any suggestions here!

  • meerkatmreow

    What evaluations have you done to assess the accuracy/robustness of your model?

  • SteveRepsold

    • 2015 FACFC Finalist

    For whatever it’s worth (certainly not statistically significant lol but…), I had a lot of success using your model as the primary basis for my research yesterday. Cash game model was more successful than the GPP, but that’s to be expected (and the GPP didn’t do too badly either).

    And it did all that in a lot less time than my research normally takes.

    Thanks!

  • Traeko

    Projected Ownership Analyst

    • 2014 RG Season Long Champion: MLB

    • 2018 RG Season Long Champion: NBA

    Looks awesome Cruzin, great job.

  • Cameron

    RG Co-Founder

    • 2014 FanDuel NFL Survivor Champion

    • 2016 RG Season Long Champion: NFL

    Can’t wait to see the ODB Model.

    Great work, Brian. Polish on.

  • nvalencia30

    • x2

      2014 FAFC Finalist

    • 2015 FAWBC Finalist

    Thank you so much for the bullpen data, with daily adjustments from Sporer. That’s a terrific aspect. Can’t wait to dig in further!

  • tone17

    I tried this out last night on the DK solo shot. Just as an experiment. It was actually my 2nd highest scoring team 4 pts behind. Only cashed for $2, but this seems like another great tool from RG. This will be part of my regular research.

  • rsigler65

    This is pretty cool. You guys amaze me sometimes with the amount of work you do and the the stuff you release for free – even if only temporary. I made a similar excel file after reading some of Bales RotoAcademy articles, but without bullpen and surrounding factors. I can’t remember for sure from looking yesterday, but do you think park factors should be included? Maybe as an option to turn on and off? I don’t know – maybe that’s already a factor in Vegas lines. I’m torn between taking advantage for myself and being wary of everyone having access! I hope these types of things are appropropriately priced in the future. Otherwise it seems like it will just even out the playing field and the only winners will be those collecting rake.

  • Cjbeast1

    Great job man! Awesome tool

  • tyrant1014

    Wow, I have nothing to say but wow in a good way,

  • tyrant1014

    would it be possible to get a breakdown for the ratngs. In a vaccum 75 is average but in baseball 75 isn’t bad unless you are talking BAA of a pitcher.

  • deactivated95162

    Wow.. very nice. will give it a try tonight.

  • aerrico26

    Very nice model. Similar to the spreadsheet that I use as a GPP only player. One thing you may want to consider adding would be a “projected ownership” factor that can be edited by the user.

  • Yukerboy

    • Blogger of the Month

    You. I love you.

  • Neiderman

    • 2016 DraftKings FBWC Finalist

    Cruz I used the model yesterday as a test on DK. It’s batting .500. For the early slate I took 5th in a single entry and finished 100 and something in the large $3 early gpp so not bad.. However they were pretty easy calls since the obvious stack was The blue Jays. I then used it in a $100 high heat for main slate which I came in last.
    So I guess my main point here is that I believe it is a great tool thank you. However for those that are concerned about helping fish.. My test last night shows you can’t just blindly pic the “CRUZ” top rated players and expect to win.

  • Yukerboy

    • Blogger of the Month

    And on the seventh day, Lord Cruz rested…

  • rsigler65

    Yeah I’ve noticed the model doesn’t seem to work as well when it’s just the picture of the baseball player…

  • RamblinMan88

    Forgive my ignorance…don’t relievers also play about 1/3 of each game on average?

  • aerrico26

    The model includes “PEN” numbers, which are the relievers for the opposing team.

  • rumsey182

    looks almost exactly like what fantasy labs has ;) wonder if that is just a figment of my imagination but cool it is free for while at least

  • SteezeMcQueen

    This is awesome!

    Also, probably an unintended consequence of the thread, but the talk about the early-season “fish” and all that definitely helps the confidence. Been playing for only a couple of weeks tops, but generally cash each night! Mainly thanks to all you guys have going on here.

    Too glad I stumbled across this site when I decided to start playing DFS.

  • SouthernShield

    I see it updated, but still listed as Aug. 31st

  • CruzinToVictory

    RG Product Coordinator

    • Blogger of the Month

    @meerkatmreow said...

    What evaluations have you done to assess the accuracy/robustness of your model?

    Actually, we have plans in this offseason to do a pretty extensive regression analysis and backtesting and make whatever tweaks that seem necessary to improve the overall accuracy of the model.

    We’ll be sure to post an announcement with all of our findings as we get ready for the full launch next season!

  • tyrant1014

    @SteezeMcQueen said...

    This is awesome!

    Also, probably an unintended consequence of the thread, but the talk about the early-season “fish” and all that definitely helps the confidence. Been playing for only a couple of weeks tops, but generally cash each night! Mainly thanks to all you guys have going on here.

    Too glad I stumbled across this site when I decided to start playing DFS.

    The MLB fish will be back on Tuesdays after the overlap of seasons start. “Terrible Tuesday”

  • fantasybender

    How does ownership play into this model? Any deductions on the GPP model for highly owned players? Please advise.

  • sangamc

    @Travisbecker said...

    Awesome! I love the bullpen aspect. I did a quick query on Baseball Reference not long ago, and found that something like 1/3 of home runs and runs scored come against relievers rather than starters.

    Yes indeed, and if the starter gets yanked early, that bullpen pitches for up to 2/3 of the game!

  • X Unread Thread
  • X Thread with New Replies*
  • *Jumps to your first unread reply

Subforum Index

RotoGrinders.com is the home of the daily fantasy sports community. Our content, rankings, member blogs, promotions and forum discussion all cater to the players that like to create a new fantasy team every day of the week.

If you or someone you know has a gambling problem, crisis counseling and referral services can be accessed by calling 1-800-GAMBLER (1-800-426-2537) (IL). Gambling problem? Call 1-800-Gambler (NJ/WV/PA), 1-800-9-WITH-IT (IN), 1-800-522-4700 (CO) or 1-800-BETS OFF (IA). 21+. NJ/PA/WV/IN/IA/CO/IL only.