Rest of Season Fantasy Baseball Projections For Pitchers (Steamer)

Our rest of season pitcher projections are updated daily. They are a combination of Razzball playing time + Saves/Holds + Quality Start projections and Steamer rate statistics. You can find MLB team rosters here.
Download Table as CSV
Note: Filters and sorting in the table below apply to the output!


Using the Grid: To sort, click the column header. Clicking once will sort in descending order. Click twice for ascending order. The Name field does not support sorting. To filter, type within the text box underneath the column headers. For text, it looks for any match (so ‘John’ under name would include Johnny Cueto and Chris Johnson). For numbers, you can use greater than and less than signs as well (so >1 in Runs would filter to only players with at least 1 Point Share in Runs). You can use multiple filters at the same time.

Methodology – Our $ methodology is based on our Point Shares (PS) – a SGP variant where we create a universe of rostered players based on the position constraints and value the difference per stat of the player vs the ‘average rostered player’. These differences are then divided by the modeled amount that represents a standing point in that format (e.g., 7 HRs, 15 RBIs). These Point Shares are then added up and divided into fixed dollar distributions for Hitters (67%), starting pitchers (usually around 23%) and relievers (usually around 9%). The sum of the category dollars + $1 equal the players total dollar estimate. $/Game represents a player’s value when he starts a game.

Roster: For ‘ESPN Roster’, this is C/1B/2B/SS/3B/5 OF/CI/MI/UTIL/9P. For ‘Yahoo Roster’, this is C/1B/2B/SS/3B/3 OF/2 UTIL/2 SP/2 RP/4 P. For AL/NL-only leagues, this is 2C/1B/2B/SS/3B/5 OF/CI/MI/UTIL/9P.

Pos / MVPos: ‘Pos’ includes every position the player is eligible. ‘MVPos’ is the position that was used for their Point Share rankings and is the position where the player is judged most valuable. The order of most valuable to least valuable for positions is: C, SS, 2B, 3B, OF, 1B, DH. Note that the position eligibility rules for ESPN and AL/NL only formats is 20 games previous season / 10 games current season while Yahoo format is 10 games previous season / 5 games current season.

Owned%: Based on ownership within the Razzball Commenter Leagues which consists of 80+ 12-team MLB leagues using the standard ESPN roster format.

Updated: | Maintained by

  1. hankp says:

    It looks like pitchers that are on/coming off the DL have no W/Ls. Garza, Beachy, Danks, etc.

  2. Nation says:

    Does someone like Shelby Miller get dinged because the perception (and correctly) that he may not throw more than 150 for the entire year?

    • I deduct some GS because he has never finished a full season. But will review to make sure I am not penalizing him too much

  3. Kyle says:

    Are guys like Ian Kennedy and Estrada really more valuable than guys like David Price, Medlen, Fister and Latos? And why? Both of those guys are available in my league and a huge list of pitchers below them are all taken.

    • @Kyle: I would say no but I do think Kennedy and Estrada are good buy-low candidates given their K/BB. I would pick them up if you have roster space.

    • @evo34: This does update just about every morning. In general, rates should match assuming we estimate the same role for a pitcher (e.g., SP vs. RP).

    • This has now been fixed as of tomorrow’s data. Ryu’s issue was a reference-related snafu (he was in as a RHP vs. LHP). This impacted a handful of rookie pitchers. Gyorko (and a few rookie hitters) were missing minor league data that impacts their projections.

      • evo34 says:

        @Rudy Gamble:

        Thanks for figuring it out. I think something happened to the formatting, though, as the raw player id tags are showing in the table now, rather than the player names as links.

        • Grey

          Grey says:

          Yeah, aware of it and working on it…

      • evo34 says:

        @Rudy Gamble:

        Just a head’s up that the rookie issue appears to still exist with pitchers as well as hitters. E.g., Ryu is 3.66 RoS projected ERA at Fangraphs and 3.85 here. Thanks.

  4. Gary says:

    Summed all pitcher wins (1490), losses (1049) and saves (589) from today’s projections. Meanwhile, there are 1374 games to be played as of this morning.

    Why is there such a disparity between wins and losses and games to be played?

    Saves have generally been one-half of wins dating back to the mid-80s. Shouldn’t saves be closer to 687? (1/2 of 1374)

    • @Gary: Steamer and I don’t force Wins/Losses to sum up to a certain number. The Wins/Losses are counting stats so heavily dependent on my playing time (IP) inputs which I don’t force per team (e.g., the Yanks must have Total GS = 162-Games to Date or Total IP = 162-Games to Date * 9).

      So the Wins total doesn’t look too off. I would’ve expected it higher than total Games only b/c I’d rather err on estimating a few more GS than too few (figure it’s impossible to forecast which SPs go down and it’s best to have some projections for the fill-in guys who’ll likely play if/when there are injuries). I’m a bit surprised that Wins and Losses have that big of a difference though so will ask Steamer what he thinks about that.

      Regarding Saves, that one’s all me. I have a few tricks on how I produce those but there is no forcing the # for a team. I’d expect those to come in below your calculation since I’ve tried to factor in some level of uncertainty with current closers and it’s too presumptuous to just take the likely remainder and give it to the other relievers on that team. Plus, there’s some regression/upper limits in my model so it won’t aim to credit more than about a 35-40 save pace.

      I’d expect if you summed up Hitter PAs or Games that you’d find similar differences.

  5. Pete says:

    Is 17 a good estimate for Kenley Jansen saves?

  6. Jon says:

    Rudy, two questions: Why isn’t Alex Wood on the ROS page? Also, why the high rating for Kevin Gausman (11.7)? I’m sure it’s a service time flukey thing for both, but just curious if you have any explanation.

    Thanks, this is such an awesome tool!

  7. Big Mike says:

    Love this tool … but why is Aroldis Chapman not listed at all anymore?

  8. hankp says:

    Rudy, Do you update the save projections in season, or do you pretty much roll with the pre season stuff there?

    • I do update Save and Hold projections daily through some automated processes and then will manually make some adjustments periodically.

      The projections will generally favor the player’s pre-season role and then ‘max’ out at a new role in about 30 days based on their stats.

      If there’s a pitcher that you feel is ‘off’ for Saves or Holds, let me know and I’ll review. I’m updating for the TB situation this morning.

  9. Big Mike says:

    Where’s Gavin Floyd?

  10. hankp101 says:

    FYI – The dollar values for the ROS pitcher projections aren’t showing up for some reason.

    • Ok been working on this today – will look into it tonight

  11. Jack R says:

    It looks like the ROS Pitching stats for Toronto players are not entirely showing up.
    H, ER, K, BB, etc.

  12. hankp101 says:

    How often do you tweak your save projections?

    • @hankp101: In April, not so much. Now there’s enough games played that I can use some code to update more dynamically based on changing bullpens.

    • @hankp101: Ok, just updated the Saves/Hold projection formula. It’s more complex than last year’s – it’s now looking not just at the last 30 days but factoring in if there’s a closer on the DL (e.g., Wade Davis won’t be getting Saves when Holland is back) and when a pitcher enters the game (more likely to get holds).

      It’s not perfect (e.g., the Red Sox have a crazy low total SV projection b/c of Uehara’s injuries and no one else has earned a save) but I think it’s pretty good.

      Let me know if anything anomalous stands out.

  13. John says:

    Are you still tweaking your projections formula for RP? I noticed you had said you think Uehara may be projected too low. Also, will there be an adjustment to Cishek’s and Holland’s projections soon? I feel like Holland is being undervalued in these projections, whereas Cishek is too high (although it remains to be seen what will happen in Miami, but it seems he is done).

  14. Aaron says:

    Cishek #9 RP??? haha

  15. Derek Holland seems to have fallen off the list. He had a pretty decent ROS projection last time I captured the data.
    First year subscriber here, keep up the good work!

    • @El Gato Grande: thanks. i think the rangers moved him to 60-day DL as a bookkeeping exercise and it screwed up my script (the day a guy is on 60 day, i assume he’ll be on DL at least 60 days going forward). It’s been corrected

  16. Nathan Lee says:

    I know we are coming up on the end of the season and all but I’m shocked that Strasburg is ranked higher than Arrieta. I have both pitchers on my fantasy team and was hoping you wouldn’t mind explaining why Strasburg is ranked higher….

    Arrieta, I think, should easily be in the top 10 if not top 5 pitchers. I understand that these are just “projections” but show the man some respect…he’s given up 7 earned runs in 70 innings pitched over his last 10 starts with an 8-1 record. Of course, Strasburg has been dealing with injuries the majority of the season and those affected a few of his starts but overall he’s pitched 91 innings for the entire season, 100 less than Arrieta, and has given up 1 more run than Arrieta. He’s ERA is a little more than double of Arrietas and the only stat he has on Arrieta is his K’s per 9 innings…which is only .4 better.

    Of course, I come here for advice to try and manage to somehow make it through the playoffs with the worst team that made it in and I end up trying to make sense of why my best player is ranked 11th for the rest of the season projections. Also, I think Tolleson should be ranked higher than David Robertson…but I won’t spit out more stats for you…I’m mainly just curious if you can explain the Arrieta/Strasburg situation, I’d appreciate it.

    And I didn’t even think about it until now but, future matchups, not sure what effect they may have on the projections, in any.


    • I agree. I wouldn’t trade Arrieta for Strasburg right now – even in a keeper league. But the projections are driven by Steamer’s methodology which has proven to be the best system out there right now. I think matchups and health play a huge role now.

      • Nathan Lee says:

        @Rudy Gamble:

        Best system out there huh?

        I don’t think matchups matter when it comes to Jake Arrieta this season…

        Go Cubbies!

  17. Sol says:

    Hi Rudy. Just starting my draft research and noticed that it looks like a bunch of RPs are missing from the projections. Just a heads up. Thanks for doing this.

    • Please look at preseason projections instead.

  18. Kevin S. says:


    The quality start projections look to be off. The projections appear to have dropped across all players? Last time I pulled the data it was showing a QS% of about 70% for the elite SPs. Now, it is projecting at around 50%.

    One Example: Kershaw projections: 29 GS, 16 wins but only 13 quality starts

    • @Kevin S.: Good catch. I reworked the QS estimation code. Now it’s coming back at about a 0.65 W: 1 QS ratio for SPs which about matches 2015 (0.68 for pitchers with 1+ GS and 0 Games relieved).

  19. Chris says:

    I am in a head to head points league and I use your rest of season projections quite often when proposing / evaluating trade offers because your projections are much better than every other site out there so I am curious how often are the rest of season projections updated?

    • We update every morning and throughout day. Playing time adjustments include DL. Toughest things are uncertain playing time situations like saves/holds.

  20. Clint Driftmier says:

    I am trying to make a run at a title… Have a trade on the table and can choose between Samardijza or Ian Kennedy to get back. Who would you take?

  21. Kevin S. says:

    Ken Giles not on this latest update

    • @Kevin S.: it’s b/c he’s marked as inactive (paternity leave this weekend) and i ignore inactive (but not DL’d) relievers since they are predominantly minor leaguers. he’ll pop back in on Monday.

  22. Damon says:

    FTI: The drop down menu from the stats tab still says 2016 Steamer/Razzball rest of season projections instead of 2017.

  23. Luke Malone says:

    Hey Rudy,

    Your site and projections are amazing, thanks for all your hard work!
    Any reason you don’t list SIERA for your “rest of season projections” as you do in your preseason projections? It could be very useful. Thanks!

  24. Ryan says:

    Alex Wood projections for G/GS are incorrect. Steamer seems to have gotten them right.

  25. Coryw says:

    Can you provide historical/previous day projections?

Leave a Reply

Your email address will not be published. Required fields are marked *