Australian Open (Men): The –20.1% Yield that Shows Betting on Underdogs is a Disaster

Well, 2 days of competition in Melbourn and 43 main-draw matches (men) have been played so far.

34 favorite wins and 9 underdog wins.

+12.3% yield if you had bet on all favorites, and -36.8% if you had done the same on all underdogs.

You might be thinking… this is a very small sample. And you are totally right: 43 bets are nothing, very small and you cannot draw any conclusions as it is not statistically significant at all. But let me use this data to introduce this article 😄.

Long-Term View: Australian Open 2010–2025

This analysis is about the performance of favorites and underdogs at the Australian Open.

Is there any clear long-term bias? some interesting stuff for bettors?

If you have been following me for some time… bingo. You already know the answer 😉. There is.

For this analysis, I have taken all the Pinnacle closing odds in every Australian Open match from 2010 to 2025. I assumed we have bet $100 on every underdog and $100 on every favorite. What would the result have been?

I've gathered a total of 2,008 valid matches/bets in this period. Some odds might have been missing, and I discarded matches where the final odds of favorite and underdog were exactly the same.

And these are the results:

If we had bet on all favorites at Pinnacle’s closing odds, we would have won around $345 and the yield/ROI would have been almost flat (+0.2%).

However, if you had done the same on the underdogs, we would have lost $40,400!, or a yield of –20.1%.

Well, wat’s clear is that betting on underdogs at the Australian Open has been a losing strategy. A disaster, actually.

Breaking Underdogs and Favorites into Groups

But let’s go one step further. You might be thinking that perhaps small underdogs are not the same as big underdogs or longshots. And you are right.

I've divided these 2,008 bets into 3 equal groups (669, 669, 670 bets), ordered by odds, to see if there is any relevant difference.

The odds ranges are defined by splitting the sample into the 3 equal groups.

We can see that longshots are the ones with the worst results (–45.8%!!).

“Medium” underdogs yielded –10.8%, and smaller underdogs –3.7%.

So it’s clear: longshots produce the most negative results. In other words, high-odds bets at the Australian Open have a very negative expected value. And the lower the odds, the lower the negative yield.

And What About the Favorites?

These are the results for favorites, also split into 3 groups:

These results are logically inversely linked to those of the outsiders.

You can see that the strongest favorites are the ones that have produced the most positive results (+1.7%).

Why This Happens

The reason for the bad performance of big underdogs is mainly the Favorite–Longshot Bias, which I explained in detail in this article.

Also, you can find useful this other post, where I analyzed the evidence of the FLB in Tennis, with 40k matches across all ATP tournaments since 2010.

Pinnacle’s and other bookmakers’ margins are highest on long odds. In Grand Slams, the gap between favorite and underdog odds is at its maximum, so this poor behavior of underdogs is quite logical.

However, the Favorite–Longshot Bias is not the only reason for this poor performance of the underdogs in the Australian Open. In other words, if we take similar odds ranges across the whole history of the ATP, the behavior of underdogs is much less negative.

There is an additional factor, which is very likely related to the fact that these are best-of-five-set matches, and the market and bettors are not fully pricing this reality.

Conclusions

Does this mean you should never bet on underdogs at the Australian Open?

No. But you must select them very carefully, because the general trend for underdogs is clearly negative — you are swimming against the current. If you bet on too many dark horses, you are very likely to lose.

It’s like buying a stock when the whole market is crashing. Yes, sometimes you can win and your stock can go up, but normally it will go down. The effect is not that intense — in a free-falling market it’s very unlikely for a stock to rise — but I use the analogy so you understand the idea.

Of course, there can be some years where underdogs can perform well overall. But most of the years this won't happen.

On the other hand, since the yield on favorites has been positive, should you blindly bet on all favorites at the Australian Open?

The answer is also also No. A +0.2% ROI is very close to zero. Basically, it’s like flipping a coin.

But is true that, betting blindly, you would probably get a long-term ROI close to zero (based on the past — we can’t guarantee it will stay that way). So, if your selection of favorites is good, if you have skill at choosing them, then it’s more likely that you will win. And the lower the odds, probably the better.