Given that stats should be used in context with one another, do you have a go-to metric to evaluate hitters? What about pitchers? Is there a stat you see misused?
Jeff Boggis (Fantasy Football Empire, @JeffBoggis): My go-to metric to evaluate hitters is Weighted On-Base Average (wOBA). This statistic values certain hits more than others and I use this stat to gauge overall player value. This stat varies by position, so I only use this stat to compare players for each position. The calculation for wOBA is wOBA = (0.690×uBB + 0.722×HBP + 0.888×1B + 1.271×2B + 1.616×3B +
2.101×HR) / (AB + BB – IBB + SF + HBP). My go-to metric to evaluate hitters is Expected Fielding Independent Pitching (xFIP). The calculation is xFIP = ((13*(Fly balls * lgHR/FB%))+(3*(BB+HBP))-(2*K))/IP + constant. I use this metric to measure how well a pitcher has been throwing. I find it more accurate than using ERA for predicting future performance.
Michael Rathburn (Rotowire, @FantasyRath): Being that I’m a Bill James nerd from back when I was 13 years old, Runs Created has always stuck with me. Now its wRC and wRC+. Weighted Runs Created and Weighted Runs Created Plus look at Similar to OPS+, Weighted Runs Created Plus (wRC+) measures how a player’s wRC compares with league average after controlling for park effects. The formula for wRC+ = (((wRAA/PA + League R/PA) + (League R/PA – Park Factor* League R/PA))/ (AL or NL wRC/PA excluding pitchers))*100
Rudy Gamble (Razzball, @RudyGamble): I do not. I look at the same basic stats that most others do (K%, BB%, SB, BABIP). I put my faith in the tools/projections/player values we’ve built at Razzball. In case that sounds egotistical, the rationale behind it is that I am not very successful at analyzing hitters/pitchers based on advanced stats. I try to put my flawed, biased thumb on the scale only in tossup situations.
Jason Collette (Rotowire, @jasoncollette): I’m a big fan of using the wOBA-xwOBA (for both hitters and pitchers) at StatCast. It is not a great predictive measure of what is to come, but it can show us players that have been underperforming or overperforming to-date. It helps put a better framing around who has been “lucky” or “unlucky” and explain variances in BABIP, HR/FB, etc.
Grey Albright (RazzBall, @razzball): For pitchers, I like using K-BB, obviously doesn’t tell us everything, but if a guy is striking out a lot of hitters and not walking many, chances are he’s going to have more success vs. less. “More success vs. less” is also a good stat. It’s where you look at Success vs. Little Effin’ succeSS. Another stat that is apparently underrated, that Archie Bradley showed us this week is the HWC — that’s Hours Without Charmin. Apparently, the higher the better. Hitters are harder to figure out, not because we have no known cases of them pooping themselves, but pitchers control the action. If a pitcher throws four balls four feet outside the zone, no hitter is going to hit it — though, some may try; hello, Dee Gordon. For hitters, I try to put eyes on them. When not possible, you have to rely on a combination of BBs, Ks, Speed, BABIP and SLG, then dig into each number to see if there’s support for each. For unstints (how I spell it), you could’ve looked at Jose Martinez’s stats coming into this year and thought, “Okay, he’s got no power, had a high BABIP and Ks a bunch,” but if you put eyes on him, you would’ve saw he’s so much more, and how I ended up drafting him in Tout NL-Only for $8, while someone like Justin Bour went for $18.
Doug Dennis (BaseballHQ, @dougdennis41): like Jason, I look for who is lucky/unlucky to help me guess ahead at recency bias. wOBA-xwOBA is as good as anything else for this. I tend to use xERA instead of xFIP, but they are very similar. I deploy a silly strategy, so often I am looking for good fits with that strategy.
Lawr Michaels (CreativeSports2, @lawrmichaels): I am pretty adamant, and simplistic about this. For hitters, OBP is what matters as if a batter cannot get on base, he cannot drive in runs, score runs, steal bases, etc. For pitchers it is WHIP for if a hurler can keep runners off base, chances are he will be successful. I do also look at K/IP and K/BB for pitchers, and BB/K and BB overall for hitters.
Gene McCaffrey (Wise Guy Baseball, @WiseGuyGene): I think the use of one number for any player, hitter or pitcher, is completely wrong-headed. At best, one number is convenient shorthand, good for evaluating trades and comparing players across eras. For our purposes, I want as many numbers as possible. Why choose a stick figure drawing when 3-D holograms are available?
Tim McCullough (Rotoexperts, @TimTenz): For both hitters and pitchers I try to stick with the skill metrics (K%, BB%, Sw Strk%, Contact rate) first then move into stats like wOBA and wRC+. I think BABIP is abused by a lot of players looking for regression and betting future rebounds or backslides. I try to look at as many metrics as possible (within reason) before making any judgments – and even then, I’ve learned to take it all with a grain of salt.
Ron Shandler (RonShandler.com, @RonShandler): What Gene said. I try to avoid most overall skills gauges as players are a composed of many different skills which usually cannot be summed up in a single number and maintain any value for fantasy roster management. I also try to avoid run-on sentences.
Phil Hertz (BaseballHQ, @prhz50): I play in very deep leagues, so for hitters I generally just run the numbers for the last 14 days and see if anything jumps out at me. For pitchers — there are generally a lot more options involving pitchers — I look at four things: BaseballHQ’s BPV numbers; K/9, Strikeouts to walks, and once again the numbers for the last 14 days.
Ray Flowers (Fantasy Guru Elite, @BaseballGuys): I try to look at the overall game, certainly beyond the “fantasy numbers” many lean on as it’s more about the process than the end result in many instances. Weighted on-base average (wOBA) and Weighted Runs Created Plus (wRC+) are two metrics one should analyze players without. I also agree with Lawr that OBP is important, as is a batters BB/K ratio. On the pitching side I look at walk rates, strikeout rates, ground ball rates, swinging strike rates and first pitch strike rates.
Mike Podhorzer (Fangraphs, @MikePodhorzer): I don’t have a go-to metric to evaluate hitters, but if I was forced to choose just one, I might go with Statcast’s FB/LD exit velocity. The metric is a measurement of how hard a batter has hit his fly balls and line drives. You can’t fake hitting it hard. The metric is a major factor in projecting home runs, as a home run typically requires a minimum exit velocity to reach and jump over the fence. Diving deeper, my own developed metrics, such as xHR/FB rate and xBABIP are the best we have available right now for backwards looking evaluations of those stats. For pitchers, I care most about SIERA, as it strips out most of the luck involved in ERA and is a far better predictor of future ERA than ERA itself, or any of the other expected ERA metrics. What’s misused? FIP, for sure, as it reflects the pitcher’s actual home runs allowed total, which is skewed by randomness. SIERA is superior.
Patrick Davitt (BaseballHQ, @patrickdavitt): Hard-hit % and contact rate for hitters, K/bb for pitchers. I also like the HQ Net Positive Outcomes metric for both. But like most of the others, I’m not married to any one metric.
Glenn Colton (Fantasy Alarm, @glenncolton1): I try to avoid focusing too heavily on any one stat or group of stats. That said, for pitchers I tend to focus on swinging strike rate as I just like hit and miss stuff. As to hitters, I like to focus on BABIP and HR/FB to try and reduce the effect of good and bad luck. Of course, the SMART system and Rules of Engagement govern all we do in fantasy baseball
Mike Gianella (Baseball Prospectus, @MikeGianella): For hitters, I don’t have a go-to metric. Aggregated hitting metrics all have their uses but because stolen bases are such a significant component of Rotisserie, none is ideal for fantasy. For pitchers, Baseball Prospectus’ DRA (Deserved Run Average) is very useful for determining a baseline for pitcher skill, although park factors, catcher framing and manager usage (among other factors) need to be considered. There isn’t an advanced metric or stat that is “bad”; it is the misapplication of these statistics to fantasy that is often at issue. Most of these metrics have their uses but are often broadly applied or – worse – used as a one-word catch-all without any caveats or explanations to their limitations. A list of players who have the best xwOBA over the last 30 days offers entertainment value but little else (if you haven’t read Jonathan Judge’s piece at BP on the limitations of “x” stats do yourself a favor and do so immediately).
Scott Swanay (FantasyBaseballSherpa, @fantasy_sherpa): For hitters I’m as big a fan of Statcast data, but overlooked in all the talk about exit velocities and launch angles is that the one stat that generally separates the best hitters and worst hitters receiving everyday at-bats is walk rate. Now walk rate might not be a sexy stat, and there are exceptions to the rule (e.g.- Jose Altuve), but collectively the top hitters have a walk rate (12%) that’s roughly twice as high as the bottom hitters (~6%). Intuitively, that makes sense. On the pitching side, more predictably, it’s strikeout rate (K/9). Batting average (against) on balls in play varies a lot less among pitchers than among hitters, so it makes sense that the pitchers who strike out a higher percentage of the hitters they face would be more successful.
Justin Mason (Friends with Fantasy Benefits, Fangraphs, Fantasy Alarm, @JustinMasonFWFB): I tend to look at stats as pieces to a puzzle. Usually in order to see the entire picture you need to examine all the pieces. What may be an important stat to one hitter’s or pitcher’s value, may not be as important to another.
Perry Van Hook (Mastersball, @): Call me “old school” (you won’t be the first), but when looking at free agent hitters, I want to look at their stats to date; teams’ lineup and usage patterns; and schedule for the coming week – no sense in adding a lefty masher like Steve Pearce or Danny Valencia if their teams are scheduled to face six RHP in the week you must have them active.
Jeff Zimmerman (Fangraphs and Fantrax, @jeffwzimmerman): With hitters, my single stat is now OPS. It’s readily available and I’ve found the overall and platoon “suck” thresholds which may cost a player playing time. When I look for hitters improving, I’ve only found plate discipline (K% and BB%) and launch angle (GB%) matter. The rest is noise. For pitchers, it’s K%-BB% with a look at GB%. Strikeouts minus walks will get an owner to 90% of a pitcher’s value. Some analysts may point to the three stats and recommend xFIP. I find xFIP flawed in that it doesn’t correctly reward those pitchers on the batted extremes who consistently have an ERA lower than their ERA estimators. I’ve created pitch ERA (pERA) to help deal with these shortcomings.
Stephania Bell (ESPN Fantasy Sports, @Stephania_ESPN): Like many folks here, I try not to rely too heavily on any one stat. OBP is a general favorite because scoring requires opportunity and what better opportunity than a hitter who is able to regularly get on base. Another worth looking at is chase rate. Despite the potential variability in strike zones that can ultimately influence this number, a hitter who is willing to lay off of pitches thrown outside the zone has a better chance of being in a favorable count (and therefore a better chance of seeing a favorable pitch and making it count). On the flip side of OBP, for pitchers, low BB/9 and favorable swing and miss percentage are attractive stats. Keep the hitter from gaining opportunity to score.
Todd Zola (Mastersball, @toddzola): Since I come up with the questions, and I prefaced this week’s with, “Given that stats should be used in context with one another”, it should come as no shock I’m in the aggregate group, looking at more than one at a time. That said, within our individual evaluation realm, we each have our own favorites, which was the purpose of the query. I must say, my colleagues came through big-time.
Most of my analysis begins with contact -for hitters and pitchers. A batter can’t be productive if he doesn’t hit the ball. The less contact a pitcher allows, the less chance there is of something detrimental occurring. So, I start there then branch out in context with what I’m looking at.
While I don’t mean to ruffle feathers, there are some issues with some of the metrics identified by my brethren. Knowing how to apply a stat is integral. Is it actionable over the given sample? Further, knowing when not to use a metric is just as important as understanding when it’s apropos.
An example is weighted on base average (wOBA). In a nutshell, wOBA is a meat grinder number designed to capture a player’s overall production potential. Mike G. hits on this – it doesn’t account for speed, obligatory in fantasy evaluation. In addition, it’s not park-corrected, which doesn’t unto itself dampen it, you just need to understand that and apply properly. Finally, wOBA is a rate stat, based on league average run-scoring potential. As such, it doesn’t incorporate team context or spot in the batting order. Identical wOBAs for two different players does not portend the same level of run production.
Another example is BABIP, especially with respect to sluggers. The formula doesn’t include homers, which are often well-hit balls. Many times, a slugger will sport a low BABIP so he’ll be labeled unlucky. The reality is, some of his hard-hit balls that would stay in the yard and be part of BABIP for other players aren’t represented for the slugger. The slugger’s BABIP isn’t always unlucky; it just doesn’t accurately capture the slugger’s profile. Again, this doesn’t make BABIP bad; it just means you need to really understand it and it’s utility.