About a month ago, I wanted to test whether a seemingly safe betting strategy would work. It was simple: Bet on “aces.” Because aces win games–that’s how they got to be aces. If you take the Kershaw’s and the Arrieta’s and the Sale’s in every game they pitch, they should win more games than they lose; and as long as the moneyline odds aren’t incredibly lopsided, they should win enough to make betting on them a worthy endeavor.

Before I go any further, what exactly is an “ace” anyway? It’s certainly up for interpretation, but I defined an ace as someone in the top 20 in WAR among starting pitchers for the 2016 season. I figured this was a better method than arbitrarily picking pitchers who I *considered* to be aces–if I did that, the analysis would have been biased. For the record, the only exceptions I made were for pitchers who had the requisite WAR to be ranked in the top 20, but had been out injured for portions of the season and therefore fell short of the required innings limit (e.g. Zack Greinke, Gerrit Cole).

My theory before I began keeping tabs on baseball’s best pitchers was that they would win close to 60% of the time. After all, that’s roughly the same win percentage of the league’s best clubs like the Cubs and Nationals. But here’s where it starts to get interesting. Aces, as you certainly would’ve guessed, are almost always favored, which means that you will have to bet a greater amount of money in order to win a certain amount. For example, in Chris Sale’s start yesterday against the Marlins, the White Sox were -140 on the moneyline, meaning you would have to bet $140 in order to win $100. This extra money placed on the betting line is also called “the juice.”

Since my hypothesis was that aces would win 60% of the time, the *minimum *average moneyline for aces needed for me to break-even if, say, I bet to win $100 in every one of their starts is **-150**. Here’s how this works numerically. If over the course of 100 starts, aces were to win 60 of those games, I would win $6,000 ($100 * 60). However, in the 40 starts lost by aces, I would also *lose *$6,000 ($150 * 40) due to the “juice.” Again, this is all assuming that the average moneyline for aces is -150.

What I came to learn, though, is that in the 92 games I looked at from July 15th to August 14th, the average moneyline total was -163. In this case, winning 60% of the time isn’t good enough. Using the same procedure as before, a 60% winning percentage would lead to a loss of $520 (($100 * $60) – ($163 * $40) = -$520). In order for this strategy to simply break-even, I would need aces to win at the same rate as the Chicago Cubs are this season, which is roughly 63%.

But let’s keep something in mind: -163 was just the *average *moneyline total for aces. For instance, in 24 of the 92 games I looked at over the past month, the moneyline total exceeded -200. If we look only at these subset of games, then aces need to be far more successful: Only winning 60% of the time would result in a *$2,000* loss if I bet to win $100 on each game (assuming the moneyline is -200). Again, in order to just break-even–forget about making a profit–aces would need to win a whopping 67% of their starts.

Of course, one of the main assumptions I’ve been working with throughout this analysis is that aces win at a 60% rate. In turns out that is wildly untrue.

So now it’s time to unveil the numbers! And boy are they ugly! Over the past month, **aces were only 41-51 (.445 win percentage)**. Hypothetically, if I bet to win $100 in every one of their 92 starts, this would have resulted in a massive $4,350 loss! In contrast, if I had bet *against *the best pitchers in baseball over the past month, then I would have been up $3,455.

Just to prove to you that I am not fudging the numbers, let’s quickly run through the same analysis as before: Since the average moneyline was roughly -163, if we multiply the 41 victories by the $100 we would have received for each of those wins ($100 * 41), we arrive at a total of $4,100. Then, if we multiply the 51 losses by the average moneyline of -163 ($163 * 51), we get $8,313. Subtract the two figures ($8,313 – $4,100) and we at least get something very close to the cumulative total I had above.

A couple caveats. First, I’d be remiss if I didn’t mention that Clayton Kershaw, who is out injured, did not factor into these findings. I’m sure if he pitched, these totals may look *a little *better. And second, this is only a month’s worth of data. This simply may have been a bad month for the best pitchers in baseball. More of an investigation, perhaps over a full season’s worth of games, is in order.

Nonetheless, there are still some interesting conclusions to draw from this analysis. Here they are as follows.

#### Aces were *consistently* unreliable

It wasn’t like there were a couple horrific days that distorted the final total: Betting on aces resulted in a loss in 24 of the 30 days I looked at.

#### The very best aces were more risky

Whenever pitchers like Max Scherzer or Madison Bumgarner took the hill, they almost always received absurdly high odds (Scherzer, for the record, had an average moneyline of -225 in his five starts during this span). This stands to reason: You’d think pitchers of this caliber would be unbeatable. Yet two striking things stand out: Pitchers who were ranked in the top 10 in WAR at the time of their specific start were only 16-26 (.380). Compare that mark to the pitchers ranked 11-20, who were an even 25-25. Moreover, when the moneyline was greater than -200, those pitchers were only 10-14 (.416) and betting to win $100 on each of their starts would have resulted in a $2,265 loss.

Maybe it was just a bad time of year for the game’s most elite starters. Yet even so, it goes to show how risky it is to bet on heavy favorites in baseball.

#### Aces were overvalued at home

Aces fared a little better at home than on the road in terms of win-loss record (.477 win percentage vs .416). However, in terms of profitability, aces lost more money at home (-$2,605 vs -$1,785). This lends credence to the idea that home-field advantage, from a betting perspective, is overrated because the prevailing logic that it is harder to win on the road than it is at home inflates the betting lines.

#### Aces struggled against other good pitchers

I noticed that when people bet on aces, there tends to be a Halo effect. Usually, all anyone notices when they look at a matchup between, say, the New York Mets and Washington Nationals is that Noah Syndergaard may be pitching–*Oh look! Syndergaard’s pitching today- put another W on the board for the Mets!* But what people typically discount is that someone capable like Tanner Roark might be pitching for the Nationals. In short, when aces faced an opposing pitcher that simply had a WAR above 0.5 (that is, they were an above-replacement level pitcher), aces were only 23-35 and would’ve lost $2,815.

#### Even against crappy pitchers, aces still weren’t profitable

Aces were more likely to win if they faced a pitcher with a WAR of 0.5 or below. However, they still wouldn’t have been profitable to bet on, as their .530 win percentage didn’t come close to breaking-even (the hypothetical loss amounted to $1,455).

#### On the bright side, underdog aces showed some promise

At least when aces were underdogs, they essentially broke-even in the thirteen times this situation occurred. However, their record was ultimately only 6-7.

### Overall, it is not wise to bet on aces

As I said earlier, I’ll probably need to follow up on this investigation at some future point. But in addition to the points I just made, there are two final takeaways. First, given how aces are almost always favored by enormous amounts, it’s clear that they need to perform at extremely high levels in order for it to be worthwhile to bet on them consistently. And second, as the last month of play has shown, it’s unrealistic to expect them to, either.

*Syndergaard photo courtesy of Getty Images*