Maintaining a positive yield over a long period, such as nine months, is a challenging task. Therefore, it’s no surprise that our overall yield across all prediction types is negative. We used a fixed stake of $1 for each match, regardless of confidence levels.First, let's look at how our overall predictions performed in 2024. The odds-based predictions, which rely on the probabilities implied by betting odds, recorded the smallest negative yield at -1.69%. To be honest, a yield of -1.69% isn’t terrible, especially given the extended period over which it was calculated. For WTA matches, however, the yield was -4.44%, highlighting a significant difference in predictability between ATP and WTA matches.The form-based predictions had a yield of -8.51% for ATP matches. Surprisingly, WTA matches performed slightly better, with a yield of -5.49%. Just as a reminder, we calculate form by assigning points for strong performances without subtracting points for underperformance.In summary, ATP matches appear to be more predictable than WTA matches. This is largely because there were fewer upsets during the season. ATP favourites, regardless of their rankings, tend to hold their advantage better than WTA favourites.

Clay emerged as the most predictable surface of the 2024 season. For ATP matches, odds-based predictions achieved a positive yield of 1.32%, while WTA odds-based predictions recorded a negative yield of -2.29%. Form-based predictions for both ATP and WTA matches on clay were in negative territory, with yields of -5.39% for ATP and -6.84% for WTA.Grass courts ranked second in predictability. Odds-based ATP predictions had a yield of -0.75%, while WTA predictions showed a slightly worse performance at -2.75%. Form-based predictions for grass matches were also negative for both ATP and WTA.When it comes to hard surfaces, results varied significantly between ATP and WTA. Hard indoor surfaces could be considered the second most predictable surface for ATP matches, though the sample size of matches was relatively small. As a result, we ranked grass surfaces higher overall. Even so, odds-based ATP predictions for hard indoor matches showed the best yield besides clay, at -0.63%, outperforming grass. Conversely, WTA matches on hard indoor surfaces had a very poor yield of -18.27%, though the number of matches was very limited.On hard outdoor surfaces, the yields were disappointing. ATP odds-based predictions recorded a yield of -5.39%, while WTA odds-based predictions followed closely at -5.99%. However, there was a small surprise: WTA form-based predictions outperformed odds-based ones, with a yield of -4.65%, slightly better than the -5.99% yield for odds.

We also simulated a system where the stake increased with higher confidence levels and decreased with lower confidence levels. However, this approach did not significantly improve the overall yield. Additionally, we experimented with random stakes in simulations, which yielded mixed results, showing slight improvements in some cases.In conclusion, there is no foolproof system that guarantees long-term success. On our Prediction Check page, you can see the rate of success for our predictions, but it's important to note that this rate doesn’t always align with the actual yield. Upsets in tennis are inherently unpredictable, and it’s crucial to keep this in mind when analyzing matches.We believe that studying player statistics is a valuable way to evaluate form and make informed decisions about match outcomes. However, remember that our predictions serve as guidelines or opinions, not certainties.In the upcoming days, we’ll focus on analyzing the most profitable and predictable tournaments, as well as players' form. Stay tuned for more insights!