I’m back with another exciting article on pitcher analytics. This week, I dive in (feet first into the shallow end mind you) with a little project built upon a fantastic article done by one of our finest Razzball baseball writers, @everywhereblair. Way back in 2021, our 3x @fswa finalist and @razzball wiffernaut (his word, not mine) penned a thrilling dissertation linking Japanese sanjikai with pitching Average Draft Position (ADP). With newly published NFBC historical baseball ADP going back to 2003, Blair climbed to the high platform and performed a perfect forward dive with three full somersaults and a half twist. The judges gave him all 10’s and I think you will too. The article explained why in fantasy baseball, pitchers don’t matter as much as people tell you they do (that’s the half twist). I’ll hit many of the high points here but do yourself a solid and go read his full work for yourself (Get Blair’s Article Here).
Unlike this award-winning piece by Blair though, what you are reading today will be more of a term paper. Not that I’m trying to cheat you out of good fantasy analytics or anything but come on folks, I loaded the hook with Ciara. If I titled this piece, What does Russell Wilson’s wife and pitching analytics have in common, would you have clicked the link? After the season he just had…Hell no! That said, let’s pause for a second and remember Russell and his greatest accomplishment(s).
Somewhere down the hall, I can hear Grey yelling, “Stop viewing that crap LB and get your ass back to work” so I guess my daydream is over! Back to the task at hand…
I hope you took some time between your own daydream and getting to this point in the article to have read Blair’s work in full. If you haven’t yet, let me give you a synopsis. Blair says ranking pitchers is meaningless “because we shouldn’t be looking at pitchers in a hierarchical manner in the first place. We should be looking at them in a probabilistic manner.” His article is “a critique of ADP, the aggregate ranking systems that replicate and reinforce ADP, and the industry that has developed around the commodification of rankings.”
Blair says, “ADP should be a guide only for when you can draft a player, not how valuable that player is.” If you get nothing else out of this article, well besides the connection between Ciara and SIERA (Skill-Interactive ERA) of course, this is the take-home message more than any other. In fact, it’s so important I’ll say it again with emphasis: “ADP should be a guide only for when you can draft a player, not how valuable that player is.”
Blair did a ton of research going back to 2004 and found that it’s extremely rare for a consensus SP1 to be drafted in the first round and return first round value. There were some close calls going back through the years but the last time an actual consensus first round ADP SP1 to return that first round value occurred in 2014. Clayton Kershaw finished as the overall #2 in fantasy baseball. Johnny Cueto and Felix Hernandez were SP2 and SP3 in that draft, finishing with an ADP value of 171 and 43, respectively.
Going back to 2004, only one other time did this occur. Can you guess the name? If you guessed Johan Santana in 2006, you get the prize. That’s right, Blair found that only two times from 2004-2020 did the fantasy community project and draft the SP1 at the proper value. Yuck!
Historically, we (the fantasy community) are almost always wrong in trusting consensus rankings on the SP1. Yet, like our little friends the lemmings, we (the fantasy community) continue to draft the same way year in and year out. Like Blair in 2020, I’m hoping to make a convert out of a few of you this season.
Let’s take a look at some data from Blair’s article. In the table below, he demonstrates the “hit rate” of pitchers by ADP over a 4-year stretch. He considers “a “hit” to be a pitcher who performs either at or better than their draft capital, and a “miss” to be a player who performs at one round or worse than their draft capital.
In the stock market, a 30%-40% return means you’re a genius at picking the winners. In fantasy baseball, not so much!
Blair feels that many fantasy baseball players choose to draft pitchers early because they are the “safest bet to return value” but you can see his research doesn’t back that up. He felt the movements to create aggregate ranking systems normalize outliers and provide fantasy baseball players with “rankings that actually recreate ADP rather than attempt to dissect and overcome the weakness of ADP.” The aggregate ranking system “normalizes misses instead of incentivizing properly valued hits that maximize value.” Here again, I’m in full agreement with Blair’s conclusions.
Blair also gets into the Pocket Aces and Pitcher-Heavy Strategies. He provides good data to support his thoughts, but I’ll leave that for you to read. I will say though, he found in 3 out of 4 years in his study, it would have been better to take the best two hitters available rather than the best two pitchers. Enough said!
Status check. Still with me? Hang in there, we’re getting closer to the 2023 data…
Let’s take a minute to talk about hitters. Are hitters in the first round generally a better investment to return their value? Yes. Take a look at this table where Blair charted the outcome of how many hitters vs pitchers drafted in the first round finished with first round value:
You see in 3 of 4 years, hitters provided better first round “value” than pitchers. The one outlier, Max Scherzer, who was actually SP2 by ADP behind Clayton Kershaw. Otherwise, all the other consensus first round pitchers by ADP failed to return their draft capital value.
So, what does this all mean? Well for starters (pun intended), you’d clearly be in a better position by drafting a bat in the first round rather than a SP. Don’t get me started on closers! That’s a topic for another day.
Next, we need to know which stats put pitchers in the top 10 for fantasy baseball. I covered a sampling of all the pitcher analytics in a previous piece (Get My Pitcher Analytics Article Here) if you want a refresher. However, Blair took the numbers from 2017-2020, checked them against multiple types of correlations, and found the results surprisingly simple. “The top fantasy pitchers almost always have top 10 finishes in innings pitched (IP), K-BB% and SIERA rates.” By combining this information with incoming news about player injuries, he concluded our rates of success on pitchers jumps dramatically.
Now I get to my stuff. Did I tell you I love spreadsheets? Well, I do. So, when I absorbed Blair’s information, it gave me a great idea (I think so anyway) to tinker with Excel and develop something useful for 2023 drafts. Here is the result:
There, all done. Good luck in your drafts!
Just kidding, folks. I’ll break this down into manageable chunks and reveal who some of these little dots belong to, and more importantly, what I think it means. This’ll be a team effort though. Remember, Blair found that a combination of quantitative data and qualitative news leads to success. So, I’ll do the number crunching and you follow the news. Hopefully, we all have successful drafts.
I started my “research” by looking at IP data. Specifically, I borrowed Rudy’s projections (Get Rudy’s Pitching Projections Here) and ranked the pitchers 1-823 based on the number of innings pitched. So, Sandy Alcantara (200.2 IP) = 1, Framber Valdez (194 IP) = 2, Corbin Burnes, Aaron Nola and Max Fried (191.0 IP) all = 3, and so on…all the way to some RP (5.8 IP) = 823. Here’s a sampling of what that looks like for a few names I’ll use for demonstration purposes.
Table Note: Disregard the order now, it’ll make sense later.
Next is SIERA (almost typed “Ciara” there). Like the FIP (Field Independent Pitching) and xFIP, SIERA attempts to determine the underlying skill level of the pitcher. Unlike the FIP, SIERA attempts to more accurately model what makes a pitcher successful. As FanGraphs puts it, “SIERA tells us more about the howand why of pitching.” Here’s how pitchers have been rated in this category:
I used the same source (Rudy’s projections) and performed the same wizardry in Excel (sorting is easy but that rank equation…) to come up with this table for the same pitchers.
You can see in this table, most of the pitchers are at least “Average” but there is still a measurable discrepancy in SIERA values and their respective ranks compared to their pitching colleagues. Without looking, any guesses who is projected to have the best SIERA value? Well, it’s a tie between Jacob deGrom and Edwin Diaz at 2.23 (Excellent). The worst? Nah, who really cares (Awful)!
You guessed it, more Excel. For this one though, K-BB% is not one of the projected stats in Rudy’s data. However, with the magic of the spreadsheet, we can easily calculate both the K% and BB%, then the K-BB% for all 823 pitchers that we need for the analysis. Why calculate for all 823 pitchers you ask? We’re getting to that. For now, here’s the K-BB% table for your viewing pleasure.
If you’ve followed closely, you know we now have all three of Blair’s variables. The twist I put on this is to determine their respective rankings as compared to the other 822 in the projections, instead of just looking at the analytics. This added component is important to me because if you look at Framber Valdez’s K-BB%, that ranking of 313 is going to be an important factor to consider, especially after the other two categories in which he scored well. He has the making of a “miss” as Blair discussed.
So, what do we do now? Let’s add ‘em up!
IP – SIERA – K-BB%
Here is the table that sums the rankings of each variable. For instance, from the previous three tables, note Corbin Burns’ rankings were 3 (IP), 17 (SIERA), and 13 (K-BB%). That equals (pulls out his handy dandy calculator watch from 1985) 33. The 3-variable sum for all 823 pitchers is calculated and then once more I determined the rank of each relative to one another.
You also note this table has a third column. These are the respective ADP values from NFBC drafts between January 1 – February 1. Remember I told you to disregard the order of pitchers in the earlier tables, well now you know the names are listed by NFBC ADP. This is kinda important when you plot the data.
Now we want to explore the data to see what it’s telling us. First though, let’s bring back that pretty picture (no, the other one) and relook at the plotted data now that we have more context.
This time, I identified a few of the names we’ve been tracking along the way. As you may have deduced by now, the yellow dots represent the ADP for the top 100 pitchers in those NFBC drafts. The blue dots represent the rankings for each player’s calculated IP-SIERA-K-BB% from the fourth table.
So, let’s briefly go back to Blair’s premise that “the top fantasy pitchers almost always have top 10 finishes in innings pitched (IP), K-BB% and SIERA rates.” Also, let’s recall the most important point in this whole article, “ADP should be a guide only for when you can draft a player, not how valuable that player is.” What this exercise does then merges those two ideas, extrapolated to all pitchers, and then compares a pitcher’s cumulative rank in these three important categories versus his ADP. I submit this gives us prospective VALUE (Author’s Note: if this concept takes off, I’m going to regret not giving this a more fun name).
The concept here is fairly simple. If a pitcher is below the line, he projects to have overall value since these three categories are so critical to historical results. The farther away from the ADP line, the more valuable this pitcher is at his draft position. Above the line, let someone else make the selection. Let’s take an example:
Sandy Alcantara has a current ADP as SP5 but projects as the #24 pitcher in cumulative IP – SIERA – K-BB%. Conversely, Carlos Rodon has an ADP of SP15 but projects as the #4 pitcher. Which one then provides better value?
Similarly, Yu Darvish and Framber Valdez are being drafted nearly interchangeably at SP30 and SP33 but Darvish’s projected rank of 17 is far superior to Valdez’s 62 (remember Framber’s K-BB%?). Same concept when you compare Tony Gonsolin to Kodai Senga or the player projected to have the most “value” in Andrew Heaney. Andrew Heaney not high on many draft boards, I know, but we’re looking for prospective value here.
You’ll note there are a lot more dots below the line. Your homework is to figure out who those belong to.
Don’t get me wrong, I’m not suggesting this should now be your draft list (remember I said I would do the quantitative part and you need to do the qualitative part). But it does provide a nice tool to help you make some tough decisions based on something more substantial than uniform color or cleat size…or ADP.
The beauty of this: The methodology is 100% interchangeable with other analytics you may like as well. Want to incorporate LD%, add a fourth calculation. Easy! Want to remove K-BB% and go with K/9 instead? Piece of cake! Play with it any way you like and fine-tune your draft strategy.
There you have it; the term paper is done. If you haven’t figured it out yet, my goal in this series of pitching articles is not to perform high-level calculus and eventually come up with the next new acronym. I much prefer to help remove the stigma of pitching analytics and find ways to enhance all our fantasy baseball experiences. I hope it gives you a little edge – unless you’re drafting against me of course! Let me know your thoughts below.
Once again, I want to thank @everywhereblair who provided the inspiration for this little exercise of mine. I probably wouldn’t have made it here strictly on my own. Once again, his full article can be found on the Razzball site (Get Rudy’s Pitching Projections Here)). Also, thanks to Rudy for all his work on player projections. He’s one of the best at this for a reason folks!
One last bit of exciting news. You may have noticed my last few articles have appeared in Blair’s traditional place in the Razzball rotation on Mondays. Starting next week, Blair is back in his normal spot. I’ll be shifting to a new day and topical focus (stay tuned). I’ve enjoyed these last few weeks focusing on pitcher analytics and hope I’ve helped you kickstart preparations for all your upcoming drafts. Pitchers and catchers report in a few weeks. How great is that!
Until next time…