On the Love of the High-Variance Player

This post is about my love for two specific professional players of radically different games: Baseball (Major League Baseball) and Overwatch (Overwatch Champions Series/OWCS).

I suspect that the venn diagram of people who are seriously into both the MLB and OWCS is not incredibly large, and the people who will know the players I am talking about even less so (to know the baseball player you would sort of have to be a recent Blue Jays or Phillies fan, but most pro Overwatch fans will know about Viol2t). So I will try to give all the needed context.

This requires a little bit of context about how I got into watching these things. I got into professional Overwatch because I played the game quite a bit in undergrad and naturally enjoyed watching higher level competition. It was especially nice because the older Overwatch competitive league (called the Overwatch League, or OWL) had city-based teams like in traditional American sports. This made it much easier to pick a team to root for, as I had just moved to Boston for university and so rooting for the Boston Uprising was natural. The OWL has since disbanded and was replaced by the OWCS which has a more traditional e-sports model of random organizations fielding rosters without any particular affiliation.

The problem with the new system was that the city-based system lead to a sort of natural leveling of the “playing field” between teams because there was some sort of salary or regional gap (I don’t know exactly what) that meant all the teams, plus or minus a Vancouver Titans, were competitive with one another. The thing that was specifically troublesome is that the Korean region is far, far superior to the other regions at Overwatch, and so post-OWL breakup instead of every team having a Korean core with some token North American or European players (and thus all being competitive with one another), the OWCS Korea region quickly became the only region producing serious contenders for trophies. I would say right now at any given time Korean teams make up the Top 3 with certainty.

This meant that as a North American viewer, I had to pick a Korean team of some sort to root for, even if just for the Korean qualifiers, otherwise I would be relegated to watching my favourite North American team play sub-par Overwatch in the regional tournaments and then get dominated at any international tournament. At the time (and continuing to the day I write this in mid-July of 2025) the Korean region hierarchy is pretty clearly ordered as Crazy Racoons on top, Team Falcons next, and then Zeta Division after them. CR and Falcons are both super-team mega projects combining the best talent from the defunct OWL, except something about the CR chemistry/teamplay works better so they virtually win everything. I felt weird about rooting for the clear favourite, and my hatred of the Falcons organization extended naturally from their CS:GO/CS2 roster so I picked Zeta sort of by default.

One of the Zeta players (at the time) was one of the all-time greatest players in Overwatch, named Viol2t. Despite the fact he is frequently considered to be one of the all-time greats (see his Liquidpedia page) he is also known for sporadically playing extremely poorly, often throwing away his life or powerful ultimate abilities in incredibly important fights. This made him, I think, very fun to watch - you never know if you will get the several-time MVP winner who can turn fights single-handedly with precise, calculated gameplay, or if the brain worm will take over and cause him to single handedly knock his team out of contention for a major tournament.

A proper telling of Viol2t’s history has been done in justice elsewhere, but the basics you should know is that he was for most of his career one of the best “flex supports” in Overwatch. Flex supports are a specific role that professional Overwatch teams that I will not go into depth on, but later in his career he switched to a similar but distinct “main support” role that required adapting to a new set of mechanical challenges. This is roughly when his gameplay switched from being consistently excellent to higher-variance, as he became known for giving up first kills in a fight and being suspectable to “Ajax” which refers to when a player plays the hero Lucio and tries to use their very impactful but infrequent to use ultimate ability but dies in the process of using it, which is a tremendous loss in value for the team. In American football terms it is the equivalent of throwing an interception, or dropping an easy fly ball in baseball. It is the sort of thing you expect from inexperienced, nervous, or incompetent players, not from perhaps the best player to touch Overwatch. This all cumulated when Zeta Division lost 3-2 to the Toronto Defiant in the semifinals of the Esports World Cup in 2024, where Viol2t ajaxed in spectacular fashion on the turning point of the final map in the last seconds of the match. After this, Viol2t played some more tournaments for Zeta before eventually being kicked and joining Team Vision, where he plays to this day, back on flex support, but still playing his new-aged high-variance game. It should be known that when he got kicked it was widely thought he chose to leave, which I think shows just how respected and high-value he is when he is playing well.

The next part of the story requires context for why I got into baseball. My family is pretty into sports (mostly the Seahawks, the American Football team from Seattle) but I never really got into it until I started grad school in CS. Most of the people in my lab are men, and I quickly realized men really mostly like talking about sports if you leave them to their own devices. I distinctly remember going to watch the finals of the Four Nations hocky tournament and hearing one of my lab mates and his friends more or less recite sports statistics to each other for three hours straight instead of having something resembling an ordinary conversation. I quickly realized I needed to get into a sport of some kind or I would have nothing to talk to my lab mates about. Football was out of the question because Toronto doesn’t have an NFL team, and I knew soccer and basketball would just be non-starters for me so I decided to get really into baseball. The fact that MLB teams play north of 160 games a season and tickets are often pretty cheap helped my decision. This ended up not really making sense because most people, even people fairly otherwise into sports think baseball is piss boring.

But I found that I actually genuinely quite enjoy baseball. I could give a long list of reasons, but I will spare the extraneous prose and simply say that I’ve been to a bunch of games and done reading on the sport generally. One of the things I appreciated the most was the game theory behind baseball: both from the perspective of an individual player (how should pitchers or batters approach at-bats for instance) and from a manager (how should I draft players or play them in the field). The player evaluation/acquisition part has been made quite famous by the movie Moneyball, which is worth a watch if you haven’t seen it.

But while the analytics/decision theory part of baseball is naturally interesting to me because I work in decision-making, part of baseball’s appeal is the fact that it is an activity with so much history which still affects the game. It is hard to convey how imbued the sport is in its own tradition: the traditional playing of “take me out to the ballgame” in the middle of the 7th, the ubiquity of hot dogs, the impractical uniforms, the fact that non-standard fields are not just possible but commonplace and are part of the “character” of a particular park. People do not appreciate it much, but professional baseball is one of the oldest professional sports on the planet, and the sport has underseen radical shifts in playing philosophy and culture over its many years (the NFL was not the cultural mainstay it is today during integration, for instance).

Often this traditional, weathered, storied side of baseball creates tension with the post Bill James/Sabermetrics baseball, the hyper optimized, analytical, bookish game played by quiet mathematicians in back-door offices. Again, Moneyball is largely responsible for educating the public about this conflict. But I find that in the stat-cast era, the sport has not lost its romance. The fact that the best players are more consistently found and developed, to me, does not make Freddy Freeman’s walk off homerun against the Yankees in the World Series, or the Blue Jay’s miraculous 2025 mid-season recovery spearheaded by George Springer on Canada Day against the Yankees, any less special (maybe I just like seeing the Yankees lose actually).

But there is one very notable, odd relic of the pre-analytics days in baseball, a holdout of a more nostalgic version of the sport, a vestigial organ that it is odd the evolutionary drive to succeed has not excised from the sport. This relic is the idea of a closer: a relief pitcher who is specifically designed to be played in a single inning in games where a team is leading by a narrow margin. Generally, a team will have a single dedicated “closer” who is brought out when a given team is winning by a somewhat narrow lead (usually 1-4 points), and will only pitch the 9th inning or in an overtime inning. They are usually brought on, regardless of platoon advantage (the idea you should generally play right-throwing pitchers against right-batting batters and vice versa), only bat for a single inning, and are often played for many days in a row if required.

A team having a dedicated closer is an anomaly that, by all measures, probably should have been optimized out in favour of playing closer by committee as appropriate given the workload of a team’s high relief bullpen, or playing relief pitchers by platoon in late-inning situations, or using closers in tied or down by 1 games, or et cetera. I think the Wikipedia article does a good job of giving an overview of the criticism of closers, and I also like the analysis this older sabermetrics book does on why generally high-relief pitchers should be used in more diverse situations, for more innings, less often. It is sometimes argued part of why closers exist is because of the “save” statistic, which basically tracks how many times a closer is used in a winning situation and wins the game. Nate Silver’s argument for an alternate metric does a good job of explaining the details. Basically, it is seen as “correct” to have a player with many saves over a season even if it costs you wins in expectation rather than have a more distributed high-leverage relief strategy. The kids nowadays would call this aura farming: having five relief pitchers who share saves and high-relief is much less cool than having your own Mariano Rivera, Eric Gagne, or Josh Hader with their own dedicated entrance music, themed light show, and dramatic nickname.

This brings us to the second subject of the story, the Toronto Blue Jay’s closer in 2025, Jeff Hoffman. Jeff Hoffman to me is the classic showman closer: he has a tall, imposing, profile with a gangly but threatening aura. He has a great entrance, with light show themed around the Marvel villain Thanos (inspired by his own Thanos tattoo) with coloured lights representing the infinity stones all set to music which transitions smoothly from Thanos’ theme to Kid Cudi’s “Alive”. I was only able to go to one giveaway game this season but I made sure it was Hoffman’s bobblehead night, where the bobblehead has him in a Jay’s uniform coloured like Thanos’ outfit with his glove decorated to look like the Infinity Gauntlet. He even shares a last name with Trevor Hoffman, a perhaps top 5 closer in history.

But to say Jeff Hoffman has had an up and down season would be a dramatic understatement. He has gone through a full roller coaster between brilliance (regular putting up 2 or 3 strikeout innings to close games) and a bullpen disaster (walking 5 players in a row and leaving Mason Fluharty to clean up his mess against Shohei Ohtani and Mookie Betts, giving up the most home runs out of any reliever in the Show).

Some variance in high-leverage relievers is not by itself all that unusual. All high-leverage guys get blown up frequently enough that you can cherry-pick disasterclass innings (other than the really good guys like Aroldis Chapman, Josh Hader and Mason Miller), and any MLB reliever (even AAA players pulled up just to fill the pen or position players pitching in ultra-low leverage relief in blowouts) do not have an earned run in the majority of their innings, just because so many hit balls end up as ground balls or fly balls which are a nearly free out. I have no way to prove this, but I think the odds are fairly good you could take any semi-athletic person able to somewhat consistently hit the strike zone on a arcing ball or eephus with an elite MLB level defensive team behind them would get out of maybe ~5% of innings clean (note - I previously had this at 40 which I think was a wild, wild over-estimate). This asymmetry is difficult to wrap your head around: even the very very very worst MLB pitcher in a year will, a lot to most of the time, have an inning that is better than some performances the very best pitchers in the sport will put up.

What is unusual is the odd picture his statistics and history tell versus his aggregate performance. Generally people accept that for high-leverage relievers having a strong strikeout rate and the associated needed ball mechanics (a reasonably fast 4-seamer, a strong chase and whiff rate on some pitches, and a high K/9 ratio). Hoffman is strong in these aspects: as of 9/11 2025, he’s 23/155 qualified pitchers in K/9, 40/155 in fourseam speed, and 98% percentive and 92% percentile in chase% and whiff% respectively.

However, he is abysmal in other metrics. He leads all qualified pitchers in his HR/FB ratio at 21.4%, and the number of home runs given up at 15, and is in the bottom 22 of earned runs allowed (ERA). His Fielding Independent Pitching (FIP) stat is in the bottom 10 of qualified pitchers at 9. Yet despite these, his SIERA score (a more advanced pitcher evaluation metric) is at 26/155 pitchers, fairly high and well above the next highest qualified Blue Jays reliever Louis Varland at 46, and well above his usually-offered replacement Seranthony Dominguez at 93. But, his xFIP, which is considered to usually be fairly similar to SIERA in terms of predictive power, puts him at 41, below Toronto relievers Brendon Little and Louis Varland by ~10 places.

So what are we to make of all of this? It is hard to tell, and I don’t pity John Schneider’s need to make a decision between keeping Hoffman as the closer or flirting with a closer-by-rotation set perhaps including some permutation of Dominguez, Varland, Lauer, Rodriguez, Scherzer, Yesavage, or whoever else.

Alright, so what do Viol2t and Jeff Hoffman have in common other than being inconsistent, and why is that interesting? I don’t really know. Some thinking: 1. People enjoy a bit of excitement when they watch sports, and being high-variance means there is a level of unpredictability that makes watching more dynamic. Watching a Eric Gagne close an entire year of saves is also fun, but might get a bit boring if you know he will virtually always get it done 2. It cuts to a very human, relatable aspect of sports psychology which is that consistency is hard, maybe harder than strong aggregate or peak performance (we have all made a free throw but to do it consistently is hard even for professional basketball players) 3. There is something appealing about the Jekyl and Hyde personality just sort of inherently. Maybe there is some Jungian explanation for why this archetype is interesting… who knows

I guess that’s it. Don’t know where I was going with this. I will end by saying that I hope Viol2t gets picked up by a better team than Vision (maybe a T1, resigned by Zeta, or Team Liquid? Would not hate him on a more competitive EMEA team either) and that Jeff Hoffman does not blow a save for the rest of the year. For me I don’t need to watch another nuclear meltdown inning; the excitement from knowing he has a choke latent in him is enough excitement for me, now I just want to watch him deliver the goods. But I think he’s got the mentality to bounce back.