There's you blueprint servebots

Never hit a 2nd serve.

How many times did Med unleash a fastball on 2nd today? Sure, you'll accumulate a handful of doubles, but when you have that powerful a weapon, never go away from it. If I was Opelka, I would talk to a analytics person and see what percentage of serves I would need to make to give me great odds of winning the match.

The key is being tough enough to stick to your plan when it starts to falter. Med never wavered today.
 

beltsman

Legend
Never hit a 2nd serve.

How many times did Med unleash a fastball on 2nd today? Sure, you'll accumulate a handful of doubles, but when you have that powerful a weapon, never go away from it. If I was Opelka, I would talk to a analytics person and see what percentage of serves I would need to make to give me great odds of winning the match.

The key is being tough enough to stick to your plan when it starts to falter. Med never wavered today.
Yes this is the future and will lead to rule change
 

Grizzy

New User
Using data from R4 US Open = More data points but higher calibre players

1st serve % in = 65%
1st serve points won = 72%
2nd serve points won = 47%

Total points 1884 / Total games 304 = on average 6 points a game

Someone smarter at math then me figure out what this actually means. From what I can tell, it seems like if you hit a second serve you are likely to lose the point 53% of the time. You are likely to double fault on a first serve style serve 35% of the time. You are likely to win your first serve point 72% of the time. It seems like it is a decent option for the MAJORITY of the ATP. It would be crazy to see it based off some of the giants that are winning like 80-90% of their first serves.

The biggest change necessary would be mental though. To get a player to actually treat a second serve as if it is a first serve. As we know with tennis, the mental game is a huge part of it, so I don't know if it would actually work out as well as you think, when the player has doubt in their mind about giving up free points 35% of the time, even if it makes statistical sense.

It's funny though because 72% of 65 is 47.... So maybe on average it would not make a difference, but for big servers it would be beneficial.
 

tudwell

Legend
More than half of the points he lost on second serve were double faults. He hit 9 of them but lost just 7 points when the second serve landed. That’s kind of insane.
 

Pheasant

Hall of Fame
Using data from R4 US Open = More data points but higher calibre players

1st serve % in = 65%
1st serve points won = 72%
2nd serve points won = 47%

Total points 1884 / Total games 304 = on average 6 points a game

Someone smarter at math then me figure out what this actually means. From what I can tell, it seems like if you hit a second serve you are likely to lose the point 53% of the time. You are likely to double fault on a first serve style serve 35% of the time. You are likely to win your first serve point 72% of the time. It seems like it is a decent option for the MAJORITY of the ATP. It would be crazy to see it based off some of the giants that are winning like 80-90% of their first serves.

The biggest change necessary would be mental though. To get a player to actually treat a second serve as if it is a first serve. As we know with tennis, the mental game is a huge part of it, so I don't know if it would actually work out as well as you think, when the player has doubt in their mind about giving up free points 35% of the time, even if it makes statistical sense.
Based on the numbers above, it’s nearly a wash to serve 2 1st serves vs a 1st and a 2nd serve

Take 100 2nd 1st serves
35 go out. That leaves 65 left.

65 x .72= 46.8 going for points.

It’s a statistical wash
 

James P

G.O.A.T.
Depends on how good your second serve is and whom your opponent is (their ROS specifically), but I could definitely see this tactic being more effective than traditional second serve.
 

tudwell

Legend
Zverev was doing that a lot too. Some massive second serves. I think he clocked one at 134 at one point. Wonder how many other players will begin to adopt it as a more regular tactic.
 

The Green Mile

Bionic Poster
Never hit a 2nd serve.

How many times did Med unleash a fastball on 2nd today? Sure, you'll accumulate a handful of doubles, but when you have that powerful a weapon, never go away from it. If I was Opelka, I would talk to a analytics person and see what percentage of serves I would need to make to give me great odds of winning the match.

The key is being tough enough to stick to your plan when it starts to falter. Med never wavered today.
Did it in his win over Djoker in Cinci 19 too, only better. He definitely has guts.
 

bnjkn

Semi-Pro
I think this strategy has to do with the opponent he was facing. He had to beat the GOAT, so regular second serves would never get the job done.
 

Grizzy

New User
Medvedev: (lifetime statistics)
60% first serves
75% 1st serve points won
52% 2nd serve points won

75% of 60% is = 45

Overall, doesn't seem like a good idea. But I know actual statistics and probability has more complicated math than this.
 

AceSalvo

Legend
And disrupt Djoko's rhythm. Med did not play ball with Djoko until the early 2nd set. Stuck to his plan pretty nicely.
 

ghostofMecir

Hall of Fame
Based on the numbers above, it’s nearly a wash to serve 2 1st serves vs a 1st and a 2nd serve

Take 100 2nd 1st serves
35 go out. That leaves 65 left.

65 x .72= 46.8 going for points.

It’s a statistical wash
Yes, but the double fault rates will go up when you play tighter points. The rate of getting in that ”second serve as a first serve” will go down.
 

Robert F

Professional
Medvedev: (lifetime statistics)
60% first serves
75% 1st serve points won
52% 2nd serve points won

75% of 60% is = 45

Overall, doesn't seem like a good idea. But I know actual statistics and probability has more complicated math than this.
So if he went for all first serves you assume the 2nd serve winning percentage would be like the 1st serve--75%

Out of 100 serves,
60 go in as first serves--winning 75% of them. Hence 60x.75=45 points won off first first serve.
40 remaining as 2nd first serves, 40x.60% of going in=24, 24 2nd first serves in at 75% winning chance =18
45+18=63 points won.

If he doesn't go for two first serves...out of 100 serves. Just standard play.
60% 1st serve in=60*.75= 45 points won on first serve
40 remaining 2nd serves winning 52% of them 40x0.52=20.8 (assuming no double faults)
Total points one with a traditional first and 2nd serve is 65.8

So it is pretty close and maybe a few double faults will make it closer.
 

StringGuruMRT

Semi-Pro
If I recall, Pete fired a lot of 2nd first serves when he played Andre in the 1990 USO Final. Or at least it seemed like he did...
 

MeatTornado

G.O.A.T.
Novak's all in on analytics and seemed to be hitting bigger 2nd serves earlier this year but even he started moving back away from it.
 

NonP

Hall of Fame
Never hit a 2nd serve.

How many times did Med unleash a fastball on 2nd today? Sure, you'll accumulate a handful of doubles, but when you have that powerful a weapon, never go away from it. If I was Opelka, I would talk to a analytics person and see what percentage of serves I would need to make to give me great odds of winning the match.

The key is being tough enough to stick to your plan when it starts to falter. Med never wavered today.
Another piece of gospel I've been preaching for years. I still remember this recent revealing exchange when Moose aptly pointed out that Pistol had zero UFEs apart from his 7 DFs in the '95 Wimby final and our resident dilettante jumped in with his typically clueless drivel and insisted that DFs are the worst types of UFEs, or when another self-infatuated poseur tried to lecture me on the same rookie nonsense after I posted irrefutable proof that top players are holding serve with more ease than ever. Au contraire today's players should DF more, not less.

If I recall, Pete fired a lot of 2nd first serves when he played Andre in the 1990 USO Final. Or at least it seemed like he did...
Pete actually didn't serve that big on 2nds in his 1st Slam final. The first blockbuster match where he really brought serious heat on both serves was the '93 Wimby final, after which Courier literally said "I faced two first serves today." :cool:
 

Wurm

Semi-Pro
Zverev was doing that a lot too. Some massive second serves. I think he clocked one at 134 at one point. Wonder how many other players will begin to adopt it as a more regular tactic.
In Zverev's case it's no more risky for him to blooter a second serve than it is to try and kick it. I wouldn't be surprised if he just doesn't have his second serve toss under control and has just been advised to hit whatever serve the ball toss suits.
 

ChrEa8

New User
I don't think this can be properly discussed through averages. Averages are infamous for covering practical reality in individual cases.

There are many variables here such as:
  • The dynamics between the server and the receiver. For the same server, the decision to either serve two 1st serves or separate 1st and 2nd serves depends on the quality of the returner. Sometimes it might make sense, sometimes not.
  • Court surfaces affects this also.
  • As already mentioned, the nerves come into play. Even if overall stats would say that even with more UE's, it would make theoretically more sense to just use your 1st serve, in practice, committing UE's generally makes the player tighter. And it also affects rallying.
One additional overall benefit of only using a 1st serve is that it simplifies things as you will have less serve techniques to use and consequently, you get more match-level repetitions on your 1st serve which in theory should make it stronger.

My opinion is that it might be better for many players to use 1st serve on their 2nd serves more often to surprise their opponent and also when they realize that the opponent is returning their 2nd serves too easily. But not as a default strategy. Also, players must get a feel on how their 1st serve is for the day before deciding on this. If they feel confident, then by all means they should try using it on 2nd serves too but even a little doubt probably means it's better to rely on a regular 2nd serve.
 

socallefty

Legend
I calculated how many more service points I would have won in a recent rec match if I hit only 1st serves as second serves using my serve stats from the match - the answer was 1 more measly point. So, I won’t be trying this anytime soon.
 

Hitman

G.O.A.T.
It was incredible serving, wasn't it?

Zverev did the same. At one point Zverev hit a 134mph second serve in that semi final.

Looks the future is to basically serve two first serves each time. That is the tactic.
 

TimHenmanATG

Hall of Fame
This is something which will have been statistically analysed over the decades. I don't think that players are still operating on the current service tactic because of tradition or convention.

Probability isn't always intuitive (just look at the birthday paradox, Monty Hall problem, gambler's fallacy, etc.).

Even if it were proven to be a little more statistically effective, then it's hard to imagine a player consistently deferring to this tactic on major points.
 

tudwell

Legend
In Zverev's case it's no more risky for him to blooter a second serve than it is to try and kick it. I wouldn't be surprised if he just doesn't have his second serve toss under control and has just been advised to hit whatever serve the ball toss suits.
That's a good point. I actually wonder how much Zverev's mental woes stem from his lack of faith in his second serve. If he developed a reliable but not attackable kick serve to fall back on (still going for other, bigger serves at times), would he still get as tight as he clearly does now in the big moments of slams?
 

ibbi

Legend
Right, but that level is not sustainable, and you're not going to win much with it unless you can play off the ground too. After the first set Medvedev's ground game was every bit as important to him as his serve, maybe more so. Also, as you say, you got to have the nerve, it's not just about the serve:p What use is having sniper skills if you don't have the guts to pull the trigger?
 

NedStark

Semi-Pro
I also see Shapovalov doing this at times. The problem is that, he hits DFs more often than not.

Another recent proponent of this was Mark Phillippoussis during his Wimbledon 2003 run.
 
Top