PRESSURE SITUATIONS 2 – SERVING TO STAY IN THE SET

PRESSURE SITUATIONS 2 – SERVING TO STAY IN THE SET

As with serving for a set, serving to stay in the set is another pressure situation that a player often faces in a match. Please note here that I’m taking about when the set is currently on serve, so at this situation the scoreline would be either *4-5 or *5-6. So if the server lost the game, he would lose the set, which is currently on serve.

This situation is slightly lower profile for most fans, pundits and traders than serving for a set, although effectively the pressure is likely to be the same on the player serving. They will be acutely aware that a failure to serve this game will cost them the set, and sometimes, the match.

As with laying the server when serving for the set, initially it would appear that this would be a viable entry point. However, as we saw in the previous chapter, I disproved that theory. I will again use statistics to assess whether the same is true laying the server when serving to stay in the set.

First of all it’s important to mention that the tick loss you would incur if the server held serve would almost always be less than if the server held serve serving for the set. That is because the set will still be on serve after the service game, as opposed to being finished if the server held when serving for the set. However, the tick gain for a break will often be bigger because if the player fails to serve out the set, it’s the end of the set. So on that basis it’s much more promising than laying a player serving to win the set.

Interestingly, the statistics also gave me insight that this was a much more viable entry point.

We saw earlier that in the ATP in my sample of bad servers, the mean projected hold was 67.2% - it was almost the same here at 67.1% and when serving for the set the player held 66.7% of occasions. That was almost the same as the mean and showed no edge – however when serving to stay in the set we did get a small edge. This time the server only held on average 62.9% - a 4.5% decrease from the mean.

This is fascinating because effectively the pressure should be very similar and led me to wonder why there was lower success for a player serving to stay in the set than serving for the set. I have no definitive answer for that, but the only thing I can think of is that the player serving for the set is often the ‘better player’ because they are in the lead by a break in the set, and hence would be more likely to succeed more often in serving the game out to win the set.

This disparity was also apparent in the WTA. We saw previously the mean projected hold was 56.6% for the bad servers sampled when serving for the set, and here again it was similar at 56.3%. We also saw in the previous chapter that the player serving for the set held 64.6% of the time (above the 63.6% WTA mean across all surfaces) – higher than average and way higher than the average for the bad servers sampled.

However, when serving to stay in the set, things were different. This time the bad servers held much lower, at 54.9% of occasions. This was 1.4% below the mean for the bad servers sampled and a huge 9.5% below the WTA mean across all surfaces.

Therefore we can start to draw conclusions about this strategy. Not only do servers serving to stay in the set at *4-5 or *5-6 hold serve less than the average ATP/WTA service game, the potential tick gain should be bigger and the tick loss should be smaller. It is clear that laying the server when serving to stay in the set would be much more advantageous than laying the server when serving for the set.