Using ISO and Age to Unearth Fantasy Gems
The key to continued success in a dynasty fantasy league format is to continually identify quality future major leaguers before other owners do. This often means culling through minor league statistics, reading scouting reports, and generally keeping one step ahead of the competition. I began thinking about strategies that would give me an advantage over my league mates. After a bit of unsuccessful trial and error with regression equations and correlation tables I happened upon something much simpler and more effective. I started by brainstorming a list of what I considered the most informative statistics and settled on ISO . I chose this based on personal preference towards hitters that specialize in run production at the expense of players with value dependent on speed. I also hoped that it would be more predictive – Garin Cecchini once stole 51 bases after all! I applied a point value to players with ISO in different ranges from 1 through 6, then did the same with age versus average age in that player’s league. Then I cross referenced players that exceeded my filter against their draft position and highest prospect rank at Baseball Prospectus and Baseball America. My goal was to identify players who were ranked outside the top 50 prospects but had the draft pedigree and statistics/age qualifications. From the statistics of all minor leaguers in 2010, I came up with the following retroactive information.
The process I outlined correctly identified 17 player seasons that ranged from “worthwhile to own” to “superstar” among the 36 that met my thresholds. As you can see, I opted to include “double players” as separate elements. In your analysis it should serve as verification that the player in question is a worthy target. (note: Eric Hosmer and Mike Trout narrowly missed qualifying twice along with Moustakas and Myers).
Seventeen out of 36 might not seem all that impressive, but consider, none of these players ranked inside BP’s or BA’s top 50 prospect lists prior to the 2010 season.
So, who did this miss?
A certain type of player is not going to show up in this analysis. So it’s not a surprise that Danny Valencia, Ben Revere, Charlie Blackmon, Logan Forsythe, Lorenzo Cain, and Dee Gordon were not identified.
Among players who met production and age requirements, the following failed at least one of the criteria. Alex Gordon (age), Josh Reddick (drafted in the 17th round), Michael Brantley (low ISO), Mark Trumbo (drafted 18th round), Mitch Moreland (drafted 17th round), Jason Kipnis (too old), Lucas Duda (age), Matt Carpenter (age and drafted in the 13th round), J.D. Martinez (age and drafted in the 20th round), Paul Goldschmidt (age), Kyle Seager (age), Brian Dozier (age and iso), and Khris Davis (age).
A few adjustments proved useful. Adding a second requirement for elite ISO (.200 and up) with a lesser age requirement adds Goldschmidt, Trumbo, and Gordon. It appears that while age vs. level is a valid filtering tool, it matters less at the Triple-A level (everyone halfway decent is several years below the average) and for college graduates at the lower minors (don’t penalize college players for age). I would recommend that anyone using this method place more emphasis on ISO at Triple-A and be willing to excuse older college players in the lower minors. Conversely, players with very low batting averages and moderate power numbers are identified as players of interest despite having very limited prospects of becoming major league players.
To save time, space, and sanity, I have summarized the results of the years following 2010.
Successes/Targets/notable players identified
2011: 11/34 – Jason Kipnis, Charlie Blackmon, Anthony Rizzo, Chris Davis, Paul Goldschmidt, Nolan Arenado, Christian Yelich
2012: 7/33 – Kole Calhoun, Yasmani Grandal, Javier Baez, Anthony Rizzo, Randall Grichuk
2013: 10/34 – Corey Dickerson, Mookie Betts, Kole Calhoun, Randal Grichuk, Corey Seager, Nick Williams, Joey Gallo, Delino DeShields
2014: 16/37 – Mookie Betts (x2), Ryan McMahon, Lewis Brinson, Nick Williams, Joey Gallo (x2), Marcus Semien
Other than the complete dud that was 2012 (unless you were able to snag Anthony Rizzo) there are some very impressive players. Perhaps we shouldn’t have been surprised to see Paul Goldschmidt, Mookie Betts, Anthony Rizzo and Corey Dickerson, come from “nowhere.” They were highly drafted and experienced significant success at a young age, yet somehow managed to elude top prospect rankings.
So here’s your strategy. When prospect rankings reset during the winter, these are the types of guys who are likely to climb their lists. For example, Wil Myers, Mike Trout, Giancarlo Stanton went from the back end of these prospect lists to the top-10 between seasons. The next time you hear about a Baseball America, Baseball Prospectus, or MLB update announced for the future, run this test on current minor league stats. You might just snag the next Mike Trout before anyone else knows who he is.
So who should we look at for 2016? It’s a bit tricky because we are going by last year’s stats and this year’s ranking which mean that prospect list editors have had a chance to catch up to successful seasons. Based on current rankings, I came up with the following list.
Here we see a few former top prospects in Franklin, Baez and Hicks. None are guaranteed a position but either could experience success when given the opportunity. Their presence on this list should serve as a reminder of what they are capable of if someone in your league has soured on them.
We also see a few moderately rated prospects in Aaron Altherr, Micah Johnson, Trevor Story, and Forrest Wall. These players have both real and fantasy value. Altherr had a very solid minor league career but didn’t really stand out with regard to power or speed. He is likely to be a useful player that can fill in off a fantasy bench. Wall has shown good speed and batting ability in his short career and he plays second base, desirable fantasy eligibility. Story began to climb the prospect ranks early in his professional career and, after a dreadful 2013, has cemented himself as the probable successor to Jose Reyes, a title he is likely to inherit late this season. He offers both moderate power and speed and has the added bonus of playing one of fantasy baseball’s weakest positions and in its most favorable hitting environment. Johnson, another middle infielder, has shown a strong ability to steal bases. Although the 84 steals from 2013 are not indicative of what to expect at the major league level, he should be a reliable source of 20-plus steals with decent batting ability, though his playing time will be limited due to poor defensive tools and a lack of on-base skills. All of these players should be owned in dynasty leagues that utilize a minor league bench, although Story has the best chance to be a significant contributor.
I find Peter O’Brien, Richie Shaffer, and Dan Vogelbach to be interesting as they have done nothing but hit at every level. They are also either at or close to the Major Leagues. Why haven’t they caught prospect rankers’ eyes? Their outlooks are clouded by concerns about position. Specifically, they all appear to be future first basemen which downgrades their real baseball value. Shaffer is blocked by Evan Longoria, at least in the short term, and O’Brien is a failed catcher turned horrendous outfielder and there is no guarantee that he ever fields a steady position at a major league level. Lastly, Vogelbach is widely considered to be a DH only. But none of this matters to a fantasy owner. As long as these hitters continue to produce they have plenty of value. They are very close to major league ready, providing additional potential value to competitive fantasy teams.
The remaining players are less known, which if properly evaluated, could lead to enormous profit on the cost or lack thereof to acquire. The commonality between these players is strong power potential with contact issues. There is a strong correlation between prospect success and walk rate, but poor strikeout numbers alone do not doom a prospect. In fact, when coupled with a high walk rate, there is nearly no negative impact from strikeouts. However, prospects that walk at an average rate need to limit their strikeouts to approximately less than 22 percent in order to boost their probability of success. Conversely, a very low walk rate almost certainly dooms all but the most talented hitting prospects, especially at the Double-A and Triple-A levels. For further information I would suggest reading research on the topic here:
Here are a few other players from my list:
|Player||BB Rate||K Rate||Verdict|
|Bobby Bradley||12.0% (high)||31.8% (high)||Buy|
|K.J. Woods||10.3% (avg)||30.3% (high)||Buy|
|Derek Fisher||11.8% (high)||23.9% (high)||Strong Buy|
|Tyler O’Neill||6.6% (low)||30.5% (high)||Wait and See|
|Cody Bellinger||9.6% (avg)||27.6% (high)||Buy|
|Jake Bauers||9.4% (avg)||13.9% (low)||Strong Buy|
Bobby Bradley had a monster 2015 season with the bat and is beginning to creep onto the back end of prospect lists despite the fact that he is locked into first base. He offers a ton of power and walks enough to counter his ghastly strikeout rate. If he’s going to be more than a true three-outcome player, he’s going to need to make a lot more contact as he moves through the minors. For reference, he is striking out 50 percent more than Adam Dunn did at the same level, albeit with more power.
K.J. Woods is a power-hitting first baseman who was drafted in the fourth round. His first season was fairly lackluster, he struck out too much and didn’t do much when he did make contact. For whatever reason, the next year he broke out despite playing at a higher level of competition, showing the power that made him an early draft pick. Considering the improvement from the previous season, any sort of gains in contact rate could be the precursor to a real breakout.
Despite having the pedigree of a first-round draft pick, Derek Fisher sometimes gets lost among the other elite Astros prospects. In his first professional season, Fisher acquitted himself quite well, hitting 22 home runs and stealing 31 bases, spending a large part of the year at High-A. Fisher currently has the plate discipline and raw skills to succeed even given an elevated strikeout rate. This is definitely a player to keep your eye on as continued improvement will lead him moving up prospect lists. Among these players, he offers the highest ceiling when considering reasonable outcomes.
Tyler O’Neill is a slugging outfielder in the Mariners’ minors. He was a third-round pick out of a high school in Canada, which, as you can guess, does not lend itself to extended development against top competition. Despite that, he destroyed High-A to the tune of 32 home runs and .874 OPS. The Cal League is a favorable environment for hitters (understatement) but reports on O’Neill back up his excellent power. Despite his extreme risk profile, he supposedly has an advanced approach that could lead to more efficiency as he gains experience. Improving his batting eye and cutting swing-and-miss is going to be crucial to his development into a legitimate major league and fantasy league option. Follow him closely to be the first to jump on him if these potential gains begin to surface.
Cody Bellinger, a 1B/OF in the Dodgers’ system had a very successful 2015 season, blasting 30 home runs in the thin air of the Cal League. The increased power likely came at the expense of contact, so progression into the top 100 prospect lists will depend on finding a balance between contact and power. If he is able to make the move to Double-A, maintain a large portion of his power, and cut his swing and miss, he may be on his way to stardom. If he can achieve two of those goals, he will likely become a future contributor.
Jake Bauers is a player that gets overlooked when it comes to prospecting simply because he’s unlikely to regularly provide plus power from a position that often expects it. He is worth considering due to his lower risk profile as well as his above average bat. An outcome of 15-20 HR with a .280 batting average is realistic, making Bauers is the safest investment among these players.
The players discussed in this article vary from those that are useful in any dynasty league to those that are only of interest to those in leagues that roster between 100-200 minor leaguers. Consider them within the context of your league. Going forward, use the filters as discussed to identify breakout candidates before your competitors do.
Wow, great stuff.
The first image is a little out of focus / blurry for some reason, but the second is great.
thanks, I was pleased with the results. Might try a few other stats/metrics in the future.
Altherr getting wrist surgery 🙁 out for 4-6 mo
Really great. Love the research. Thanks for providing relatively unknown guys to be on the lookout for.
I like the idea here. Two thoughts: would applying simple league and park factors help? would SLG/AVG be a better estimate of pure power than SLG-AVG?
I’ve always had an issue with ISO, as it fails to “isolate” the fact that almost every BIP has a chance to be a double or triple. So if a guy increases his BA from .220 to .260 and has no more power than before, his ISO will almost always go up — which leads people to think he has developed more raw power.
Thanks for the comment. I also don’t LOVE ISO but for a different reason. In your scenario, let’s say a hitter has 24 hits in 100 at bats with 5 of them being home runs.
AVG = .240
SLG = .390
ISO = .150
Now imagine 2 more singles
AVG = .260
SLG = .410
ISO = .150
So ISO will not go up with an increase in batting average unless it is accompanied by extra base hits. I would actually prefer the batting average to be as high as possible given similar ISO’s. By definition it’s the extra bases due to extra base hits. (15 extra bases from the HR/100 AB)
My issue with ISO is that low batting averages with just a moderate amount of power can seem elite.
Consider the same 24 hits. An ISO of .150 is kinda ok but it really shouldn’t be simply because poor slugging only looks acceptable next to the even worse batting average.
I don’t think SLG/AVG is really the answer either. It has the similar effect of rewarding a lower denominator (poor average).
Scenario 1: .390/.240 = 1.625
Scenario 2: .410/.260 = 1.577
Yet we’d both agree that scenario 2 was actually preferable. (19 singles and 5 HR vs. 21 singles and 5 HR)
Let me know what you think.
“So ISO will not go up with an increase in batting average unless it is accompanied by extra base hits. ”
My point is that a lot of soft BIP end up being doubles and triples, so an increase in hits will almost always be accompanied by an increase in XBH, even if the hitter has little power.
As for SLG/AVG, I prefer it only if we are trying to truly isolate power. I don’t think it’s a useful indicator of batter quality overall. I just brought it up bc it seemed like you were trying to assess the predictive ability of minor league isolated power. In any case, I look forward to future studies of yours, perhaps with combinations of statistics that incorporate both power and contact skills.
Just curious: of all the non-top 50 players you evaluated here, which one or two do you feel are most likely to become solid fantasy assets?
Fair enough, ISO was just the first thing I looked at and it seemed to work fairly well. I’m learning how to use MySQL and R so I should be able to quickly look at a lot more stats/metrics/etc.
I’d say I feel most strongly about Trevor Story and Dan Vogelbach simply because they should be able to hit as soon as this year if they were promoted and their current contextual situations are great. Though it’s hard to believe Vogelbach would get the call in Chicago. Micah Johnson is almost a lock to be an acceptable fantasy contributor if he can get in a roster just because he’ll be good for 25+ SB.
In the deeper group I feel Bellinger and Fisher have the combination of upside and a bit less work to do in fixing K and BB than some of the other guys.
Bauer should be a major leaguer but he might be James Loney.
yet another good read .. i love your articles , they’re very detailed and keeps the reader interested .. keep up the good work DG
So what am I missing here? I’ve used ISO to evaluate minor leaguers for the past few offseasons. My metrics? Look for the best ISO, make sure its not a career minor leaguer, and add them to my scouting list. Its part of the reason I get my hands on good prospects before everyone else (Aaron Judge, Lewis Brinson, Max Kepler, and Nick Williams for three examples).
I usually ignore the age completely, unless, like I said, its a career minor leaguer. It seems like you’ve made things more complicated than need be. Unless I’m missing something.
And I can’t count…four examples, not three examples.
The age is merely used to cut down on sample size. I think I noted that in my case study year, removing the age restriction would have only missed a few guys but at the expense of a few hundred more players to sift through. Same with draft round. Why discriminate against lower round draftees? Mainly because they are rarely successful. I just wanted a small list that you could reliably pull a few hidden prospects from.
[…] previously wrote an article in which I described a method to identify future fantasy assets before they rise up prospects lists […]
[…] Using ISO and Age to Unearth Fantasy Gems: This was my first actual bit of conceptual research, as I was not at all sure I’d even find a predictive link between minor league statistics and major league success beyond the obvious well-regarded prospects. My six deeper targets that I identified were Bobby Bradley, K.J. Woods, Derek Fisher, Tyler O’Neill, Cody Bellinger, and Jake Bauers. […]