Brecht,
about CPU, it takes more risk so the ball is a bit faster at same Power & Consistency ; also, higher Consistency raises the B1 and B2 serves' speed, even if the Power doesn't change ; lastly, CPU never puts spin in his 2nd serve, and if you put some, it lowers your serve speed significantly.
About online, I'm afraid you're very not alone in that case... That's one of the reasons that make me think to bring some changes to the Fair mode.
adnerif they were to hit 2nd serve at same percentage as 1st serve, then why do you think they wouldn't go for them?
The main reason is stated just a bit before your question : IRL, most players have around 60% of serves in, if not less. Which such %, the number of double faults would be very important if they were doing 2 risky serves.
But I can see 2 more reasons, even if they were actually getting 70-75% of 1st serves in :
1- to know if it's better to do a risked serve or a safe one for the 2nd serve, they should know what is their win % after each kind of serve ; this is something that depends of your opponent, of your form of the day, and even can vary over the match ; doing the stats & the maths in real time on their chair at each change of ends would likely be a good source of headaches
A quick explanation of what I mean ; is it any better to do a risky serve that has 75% chances to be in if it leads u to win 80% of the points than doing a safe serve that has 100% chances to be in if it leads u to win 60% of the points..? ... think ... think ... think ... Answer: it's same, except that if you go for the safe serve, you'll never do a double fault on a break point
2- if you always go for the risky serve, what do u do when u face a break point ? If your opponent is good, doing a double fault can mean losing the set ; this would greatly increase the stress even for most cold blooded players ; thus they would likely go for a safe serve ; what happens when it's the 1st time you do this safe serve since the beginning of the match or of the set ? Lacking some fresh practice on the spin & the momentum, it might not be as safe as it should, then that might even help to raise the stress a lot ; all in all, it doesn't seem much of a good strategy
From personal experience & some educated guess, I don't think the stress can lower by more than 5% the chances of the risky serve when it's done as a 2nd serve (as long as it's not against a break point). But I'm afraid that it's something I can't prove. Like the opposite can't be proven either. Some real pro testimony on this topic would be nice, but I doubt any one talked about this topic...
Oh, and thanks to let me know that IRL the 2nd serve is slower than the 1st serve, I never known that...

If u had carefully read my messages, u'd had noticed that I never stated anything in contradiction with that, so it was really not useful to expose all that links...
