PDA

View Full Version : Strategy for ordering team lineup in matches


OrangePower
02-25-2009, 11:35 PM
Our league season starts in a few weeks and we are strategizing on how to best order our lineup in matches (singles 1, singles 2, dubs 1, dubs 2, dubs 3).

Some context: we are a 4.0 team with a decent chance of making the local playoff - we're probably 2nd or 3rd strongest out of 12 teams. Also, the teams in our flight generally don't order their lineups based on strength, meaning that for example dubs1 / dubs2 / dubs 3 is just a way to establish who plays who on which court but implies nothing about the relative strength of the players in the team.

So we're deciding between two strategies:

A - Mix it up pretty much randomly each week. That way, there is no thought that needs to be put into it, and no pattern for other teams to try and read. Let the cards fall as they will each week, and may the better team on the day win.

B - Establish an order - not based on strength, but just an order that we stick to each week. That does make us predictable to other teams (in terms of who will be playing in what spot), and maybe they can take advantage of what they think are favorable matchups, but it also sends a message: we're confident that you can try match up anyway you want, and still we will beat you! Also, it keeps an ace up our sleeve for late in the season or in playoffs: when faced with a critical match, we will already have established a pattern that the other team will be expecting us to follow, and then when we need it most and when they least expect it, we have the option of switching things around if we think it will benefit us (and spoil any intended matchup they might have been targeting).

What do you all think?

raiden031
02-26-2009, 03:31 AM
Personally I think it would be better if everyone ordered by strength. I hate mismatches (top player/team verses bottom player/team) because they are often a waste of time and no fun for either team, and stacking your lineups defeats the purpose of playing an NTRP league to keep matches competitive.

heninfan99
02-26-2009, 05:04 AM
Simple: Put the players with beer guts on doubles and the thinner players on singles. :)

Pompitous of Love
02-26-2009, 05:24 AM
...we are a 4.0 team with a decent chance of making the local playoff - we're probably 2nd or 3rd strongest out of 12 teams...the teams in our flight generally don't order their lineups based on strength...who plays who on which court implies nothing about the relative strength of the players in the team.

My team is in exactly the same position (except 7 teams not 12).
We'll play a RR of each team twice. At the 1/2 way mark we're in a 3-way tie at 5&1, with my team behind a few courts, taking the 3rd place spot.

I've been employing the random strategy so far this season, but am re-thinking it.
For those (two) teams that have equal and sometimes stronger lineups, the random strategy keeps them from targeting a particular weak spot in my roster (not that any of them are really that weak). We've played the top two teams each once and we won and lost with 3-2 decisions. The random line up was employed by all 3 of us acting as captains and it seemed to even the playing field.

For the other teams that we should beat, the random strategy doesn't hurt us but sometimes squanders a strong court to less skilled opponents. This has lead to more 3-2 wins than we would want. These lower-half-of-the-league teams also tend to play order of strength lineups, and now we sit in third place where we could be farther up if we had done the same.

But line-up strategic thinking is so full of assumptions. It's really just a crapshoot. Who knows what an opposing Captain will do. Some of them you may be able to read, others not. Anybody who is willing to look at seasonal and long-term records of their opponents will have an advantage over those who don't. So my advice is to look and adapt. You're already ahead of the game by thinking about it.

But, for the fun of it, here's one way I'd play your situation:
Since you have 12 teams, I'm assuming you'll play each just once? so a RR of 11 matches.
What if you play a set line up for all the matches preceding your first big match with one of the big teams. They see that you're predictable and try to pounce on it, then you switch it up to your benefit (assuming you can see some advantage based on their record andyou can predict their response). Might help once, but probably not against the next big hitting team.

Also, one caveat (an obvious one) when playing stronger players on courts #2 (singles) and #3 (dubs). If you suspect a team might not field a full 8-man roster, these will be the defaulted courts, and you'll be really ticked that you placed your big players there. Been there...

Good luck with the season

JavierLW
02-26-2009, 05:49 AM
Our league season starts in a few weeks and we are strategizing on how to best order our lineup in matches (singles 1, singles 2, dubs 1, dubs 2, dubs 3).

Some context: we are a 4.0 team with a decent chance of making the local playoff - we're probably 2nd or 3rd strongest out of 12 teams. Also, the teams in our flight generally don't order their lineups based on strength, meaning that for example dubs1 / dubs2 / dubs 3 is just a way to establish who plays who on which court but implies nothing about the relative strength of the players in the team.

So we're deciding between two strategies:

A - Mix it up pretty much randomly each week. That way, there is no thought that needs to be put into it, and no pattern for other teams to try and read. Let the cards fall as they will each week, and may the better team on the day win.

B - Establish an order - not based on strength, but just an order that we stick to each week. That does make us predictable to other teams (in terms of who will be playing in what spot), and maybe they can take advantage of what they think are favorable matchups, but it also sends a message: we're confident that you can try match up anyway you want, and still we will beat you! Also, it keeps an ace up our sleeve for late in the season or in playoffs: when faced with a critical match, we will already have established a pattern that the other team will be expecting us to follow, and then when we need it most and when they least expect it, we have the option of switching things around if we think it will benefit us (and spoil any intended matchup they might have been targeting).

What do you all think?

If you know your opponents enough, try to match what you think they are going to do.

If your goal is to go straight up against them, guess how they stack and do the same thing. You may be wrong but it's better then just guessing or putting names in a hat. Put some thought into it.....

If they are just going random, then there is no point in you even worrying about being "predictable" because they are not even paying attention.

I only stack against certain teams (like against other teams that stack or against ones where I know our best team doesnt match up well with their best team). But it gives the affect that if anyone looks at my entire schedule when the year is over it looks like Im mixing it up a lot.

Is this league where team wins matter, or individual wins?

heninfan99
02-26-2009, 06:03 AM
All roads lead to stacking.

cak
02-26-2009, 07:26 AM
For the teams you know you will beat going straight up is a power position. There was one season I was captain and our team was so strong it really didn't matter. I played strongest at one, weakest at 3, knowing my weakest could take their strongest. It started you out with a head of confidence that tended to deflate the other team before they stepped on the court.

For teams closer to your level, the one court advantage you could get by mixing it up, and ending up with your strongest against their second strongest team, would probably be worth it. Those are the teams to go random on. Though as Pomp pointed out, be careful putting your strongest team on the 3rd doubles or second singles court. That may be defaulted.

Last season my partner and I were arguably the strongest doubles pair on our team. However she worked, and I had a teenager to drive to after school stuff, so if it was a split start we would take the later start. I would have thought at least some teams would end up with their stronger teams at the later start for the same reasons. Not so much. Those few matches were blowouts.

LuckyR
02-26-2009, 08:48 AM
You mention that most of the teams you play against randomly assign their teams. You can't stack if you don't know their lineup (hence why the random method is a great strategy).

This whole topic is yet another argument in favor of a round robin format rather than 1 vs 1, 2 vs 2 etc that there is now.

OrangePower
02-26-2009, 04:04 PM
Personally I think it would be better if everyone ordered by strength. I hate mismatches (top player/team verses bottom player/team) because they are often a waste of time and no fun for either team, and stacking your lineups defeats the purpose of playing an NTRP league to keep matches competitive.

I agree, and for the same reasons, but alas, nobody else in the league orders based on strength.

OrangePower
02-26-2009, 04:05 PM
Simple: Put the players with beer guts on doubles and the thinner players on singles. :)

Um. We all have beer guts! :)

OrangePower
02-26-2009, 04:07 PM
My team is in exactly the same position (except 7 teams not 12).
We'll play a RR of each team twice. At the 1/2 way mark we're in a 3-way tie at 5&1, with my team behind a few courts, taking the 3rd place spot.

I've been employing the random strategy so far this season, but am re-thinking it.
For those (two) teams that have equal and sometimes stronger lineups, the random strategy keeps them from targeting a particular weak spot in my roster (not that any of them are really that weak). We've played the top two teams each once and we won and lost with 3-2 decisions. The random line up was employed by all 3 of us acting as captains and it seemed to even the playing field.

For the other teams that we should beat, the random strategy doesn't hurt us but sometimes squanders a strong court to less skilled opponents. This has lead to more 3-2 wins than we would want. These lower-half-of-the-league teams also tend to play order of strength lineups, and now we sit in third place where we could be farther up if we had done the same.

But line-up strategic thinking is so full of assumptions. It's really just a crapshoot. Who knows what an opposing Captain will do. Some of them you may be able to read, others not. Anybody who is willing to look at seasonal and long-term records of their opponents will have an advantage over those who don't. So my advice is to look and adapt. You're already ahead of the game by thinking about it.

But, for the fun of it, here's one way I'd play your situation:
Since you have 12 teams, I'm assuming you'll play each just once? so a RR of 11 matches.
What if you play a set line up for all the matches preceding your first big match with one of the big teams. They see that you're predictable and try to pounce on it, then you switch it up to your benefit (assuming you can see some advantage based on their record andyou can predict their response). Might help once, but probably not against the next big hitting team.

Also, one caveat (an obvious one) when playing stronger players on courts #2 (singles) and #3 (dubs). If you suspect a team might not field a full 8-man roster, these will be the defaulted courts, and you'll be really ticked that you placed your big players there. Been there...

Good luck with the season

Exactly what we're thinking...

OrangePower
02-26-2009, 04:11 PM
If you know your opponents enough, try to match what you think they are going to do.

If your goal is to go straight up against them, guess how they stack and do the same thing. You may be wrong but it's better then just guessing or putting names in a hat. Put some thought into it.....

If they are just going random, then there is no point in you even worrying about being "predictable" because they are not even paying attention.

I only stack against certain teams (like against other teams that stack or against ones where I know our best team doesnt match up well with their best team). But it gives the affect that if anyone looks at my entire schedule when the year is over it looks like Im mixing it up a lot.

Is this league where team wins matter, or individual wins?

Team wins. Individual wins matter only as a tie-breaker in case teams are tied based on team wins.

So it's better to line up such that you have a 80% chance of winning 3/2 rather than lining up with a 70% chance of winning 5/0 but a higher chance of losing the overall team match.

OrangePower
02-26-2009, 04:22 PM
You mention that most of the teams you play against randomly assign their teams. You can't stack if you don't know their lineup (hence why the random method is a great strategy).

This whole topic is yet another argument in favor of a round robin format rather than 1 vs 1, 2 vs 2 etc that there is now.

Exactly. Stacking won't work. So that's why we will either go random, or predictable (with the idea of mixing it up when it really matters and our opponents think they have a read on us)...

And I would much rather prefer a round robin format. I once played in a league that did exactly that (although it was doubles only). The format was:

* 6 players (3 pairs) to a team
* Each pair plays 11 games against each of the opposing pairs. All 11 games are played, so scores can be 6-5, 8-3, 10-1, etc.
* Each pair plays 33 games in total, and the match consists of 99 games in total.
* Scores are recorded as number of games won.
* The team with 50+ games at the end of the match wins the match.

No stacking, no lineup fiddling, and every single game counts. Made for exciting tennis because typically boths teams would still have a mathematical shot of winning going into the final rotation.

JavierLW
02-26-2009, 04:33 PM
Team wins. Individual wins matter only as a tie-breaker in case teams are tied based on team wins.

So it's better to line up such that you have a 80% chance of winning 3/2 rather than lining up with a 70% chance of winning 5/0 but a higher chance of losing the overall team match.

Ah, in our league individual wins count.

So if it's a team that we are on par with or better then, getting a straight up matchup with them is ideal so we can get as many wins as possible.

But if it's a stronger team then we have to try to get as many wins as possible (even if it's just 1), because losing 2-3 is nothing compared to losing 1-4 or 0-5.

In your case though if the team was about your level or worse, I think the same sorts of ideas thing applys though but for different reasons.

I think what is important to the equation is this:

Singles - Put your two best players at singles if you can. If your singles players are top notch you'll get two wins a lot which makes the rest of the lineup a lot easier. In a lot of cases you'll always at least get #2 this way because a lot of teams throw just anyone at #2 just to be nice and let someone play singles. (dont be one of those teams)

Then you're only looking at needing 1 or 2 wins at doubles.

However if you do the math on it, stacking in a lot of cases doesnt help you that much because you almost never have a 100% chance of winning or a 0% chance of losing. (although it sometimes seems that way)

Let's say you only think you need one single win out of the doubles.

If all 3 positions have a 50% chance of winning or losing, and you define failure as losing all 3 then that gives you a (.5 * .5 * .5) or 12.5% chance of failure.

So then you might think that stacking 3-2-1 would be better.

So let's say for the sake of argument that increases your chance of losing at #1 to 80%, and increases your chances of losing at #3 to 20%.

Then your odds of failure are: (.8 * .5 * .2) or a 8% chance of failure.

It decreases your odds maybe (if those estimates are even right), but not by that much. (unless you play 25 or 30 matches in a season)

Also besides that consider that on most teams #1 isnt always necessarily way superior to #2 which is way superior to #3.

Some teams may have 2 very strong doubles teams and some may even have 3. So stacking is even less effective in those cases.

Also there are teams where the captain doesnt make a good lineup and doesnt even know the strength of his own players. (case in point you cant assume anything by the position alone unless you notice someone has a good record)

Pompitous of Love
02-27-2009, 05:28 AM
Good points Javier. Although the subjectivity of the percentages takes a little away from it's punch, this kind of quantitative analysis is a great way to think about your teams chance to take a match.

But I really like what OP writes (esp. when Team wins are what determines league winners):
...It's better to line up such that you have a 80% chance of winning 3/2 rather than lining up with a 70% chance of winning 5/0...

It's still way subjective (the %'s), but absolutely man! A team win is the real deal. Then there's no need to worry about Courts won/lost, or Sets lost, or Games lost...just win the darn thing!:)

Topaz
02-27-2009, 05:53 AM
In teams that I have captained or co-captained, whether or not we 'stacked' depended on #1 - our opponents, and #2 - who we had available for that match.

I'd always much rather have enough players that it really doesn't matter what court I put them on because they are strong players. On one team, we pretty much all took turns playing the different spots (and TennisLink doesn't care if it says you played #1 or #3 doubles). There are some players, though, who will resist playing certain position because of fear ('please don't put me on #1') or ego ('what do you mean I'm on court #3?'). Hopefully you won't have any of those issues!

JavierLW
02-27-2009, 09:57 AM
In teams that I have captained or co-captained, whether or not we 'stacked' depended on #1 - our opponents, and #2 - who we had available for that match.

I'd always much rather have enough players that it really doesn't matter what court I put them on because they are strong players.

That's a good point, Ive noticed when there is a team that always tends to do very well and tends to always take first, they rarely stack because it doesnt matter. They are strong at all positions usually.

(and that's under our system where every single match counts for something)

OrangePower
02-27-2009, 11:42 AM
All roads lead to stacking.

Seems like you hit the nail on the head - even this thread has turned into a stacking discussion :-|

Unfortunately stacking is only relevant if there is an assumption that at least some teams will order based on strength. And teams in my league don't...

OrangePower
02-27-2009, 11:46 AM
...
Singles - Put your two best players at singles if you can. If your singles players are top notch you'll get two wins a lot which makes the rest of the lineup a lot easier. In a lot of cases you'll always at least get #2 this way because a lot of teams throw just anyone at #2 just to be nice and let someone play singles. (dont be one of those teams)...

Absolutely.

By the way, 'best' is also relative to singles versus doubles; in our case, our two best singles players are only average at doubles, and our best doubles players are just average at singles. So that makes the decision easier. Would be harder if say we had a player who was strongest at both singles and doubles. In that case I'd agree that most of the time it makes sense to have this player play singles.

JavierLW
02-27-2009, 12:08 PM
Absolutely.

By the way, 'best' is also relative to singles versus doubles; in our case, our two best singles players are only average at doubles, and our best doubles players are just average at singles. So that makes the decision easier. Would be harder if say we had a player who was strongest at both singles and doubles. In that case I'd agree that most of the time it makes sense to have this player play singles.

Right, I believe that it's always easier to have singles players who are at least strong enough relative to everyone else that's possibly at singles out there. (I didnt want to say my "best" players because sometimes 3 or 4 guys can win at singles, but only 2 of them are good in a doubles match)

Doubles lineups are just way more complicated to me because you have to put pairs of people together that make up good TEAMS, which is more important then who's playing #1 or #2 or #3.

A solid singles position only takes one good player. That same good player can lose easily at doubles if they dont have a good partner. (unless they are say a 4.5 player playing in the 3.5 league or something along those lines)

Our league is a bit different because of the rule where individual matches count and not team wins.

So that means for most teams in most matches, it's actually stupid to stack because you need as many wins as possible.

But Ive found when playing the best team it's sometimes necessary because losing 2-3 isnt all that different from 3-2.

Anyway because less teams stack it means I think I see more order of strength then most of you guys do in your leagues. Usually captains are easily identifyed by 4 groups that Ive noticed:

1) Ones that never stack. (they have more reason not to in our league)

2) Ones that stack everytime (these guys are stupid in our league)

3) Ones that rotate their entire lineup including partners from week to week (that's how they play indoors in the winter so they dont think twice about doing it in League tennis)

4) Ones that stack occasionally for a reason such as matchup problems or as a defense against a team that is stacking on them. (this suggests though that the captain put some thought into what the other captain was going to do, I havent met too many of those. Most people just seem to blindly operate on some belief that it's either GOOD or BAD to stack and they dont pay any attention to their opponents)

goober
02-27-2009, 12:15 PM
Um. We all have beer guts! :)

Everybody lose their beer guts. Then you won't need to worry about stacking.:)

How many make it to playoffs out of 12. If it is at least 4 than it won't matter. You should make it if you are truly the 2nd or 3rd strongest team.

OrangePower
02-27-2009, 12:27 PM
Doubles lineups are just way more complicated to me because you have to put pairs of people together that make up good TEAMS, which is more important then who's playing #1 or #2 or #3.

Right. We're lucky in that we have 4 established doubles partnerships that have been playing together for a long time. So our pairings are set, and then each week we play 3 out of the 4 partnerships based on availability. We've had a few times where we've had to mix and match because 1 person from each of two different pairs was not available, but that's been the exception.

I personally can't understand why teams would ever pair players up differently every week unless they really have no choice. Some players / styles mesh well, and some just don't. Plus, knowing your partner well really helps in anticipation etc. I think having an established pairing is worth maybe 1-2 games per set right there.

OrangePower
02-27-2009, 12:30 PM
Everybody lose their beer guts. Then you won't need to worry about stacking.:)

How many make it to playoffs out of 12. If it is at least 4 than it won't matter. You should make it if you are truly the 2nd or 3rd strongest team.

Hey, we work hard at maintaining our beer guts :)

Four make it to playoffs, so you're right, we should make it anyway. I'm really thinking ahead to playoffs already... will playing a predictable lineup during the regular season give us an edge during the playoffs when we suddenly spring an unexpected order on our poor victims.

JavierLW
02-27-2009, 12:54 PM
I personally can't understand why teams would ever pair players up differently every week unless they really have no choice. Some players / styles mesh well, and some just don't. Plus, knowing your partner well really helps in anticipation etc. I think having an established pairing is worth maybe 1-2 games per set right there.

Maybe it's a location thing here.

In the winter here, many of the club "doubles" leagues are the ones where you have four guys or ladys on a court, they play X number of games and maybe a tiebreaker, and then they switch partners, until everyone has played with everyone.

Then to keep score (if they keep score) they add up how many games each individual has won.

For adults, that's pretty much the norm at least at 3.0 to 4.0 around here, and unless you play in a NTRP tournament (which there are only 3 of in the winter here), most players dont get a chance to play a real doubles match unless it's on their own. (4.5 players are out playing open tournaments or they move somewhere where it's warmer. :-) )

So the point is I think it generates a whole crop of indoor club players who run USTA League teams in the summer and they dont think twice about mixing everyone up from week to week.

I know at least one team where I dont think their players are any worse really then my team's but they take pretty much last or close to last every year.

I was even on the same guy's team at 3.0 for one season, and we had our first "practice". We had 8 guys and after every game we played we all rotated to the left so you had a different partner everytime. That was the most useless "practice" Ive ever seen.

But I guess to be fair, these teams sign up year after year which is cool.


I doubt they care much about taking first anyway (less then we do anyway), they must just be having fun somehow. Another aspect is some teams are not really about the whole "team" concept, they are really just a collection of individuals who signed up so they can play tennis, and they dont really care at all how the team looks in the standings or if the team wins or loses.

Anyway that's part of making my lineup, I try to identify those teams and take that into consideration.

For me it means Im going straight up for sure, but it also means that if I want to balance being "fair" with being successful, this is the time that I can maybe give some of my better players a rest and play the rest of my players who unlike these guys will definately win because they are not too shabby either and they have real doubles experience.

Topaz
02-27-2009, 12:58 PM
That's a good point, Ive noticed when there is a team that always tends to do very well and tends to always take first, they rarely stack because it doesnt matter. They are strong at all positions usually.

(and that's under our system where every single match counts for something)

Yes, I have found the strongest teams in our area are also the deepest in talent.

cak
02-27-2009, 01:54 PM
Personally I think it would be better if everyone ordered by strength.

If that were a rule, how would you enforce it? If you were playing against a team of 4.0 players, how would you prove they didn't play their best at position one?

JavierLW
02-27-2009, 02:08 PM
If that were a rule, how would you enforce it? If you were playing against a team of 4.0 players, how would you prove they didn't play their best at position one?

He didnt say it should be a rule.

He's just saying it would be better if nobody stacked.

It would be better if we all picked up trash when we see it laying around as well, but it's not a law or anything.

His sentiment is expressed by people who are out there to get a good match.

But in reality as far as Im concerned if you are in a 3.0 league, you cant blame stacking on why you didnt get a good match. Everyone should be competitive (at least to the point where you have to work hard, maybe some people are better at winning then others but that's not the same thing).

Ive stacked against some teams (usually the 1st place SuperTeams) and they've whined about the same thing.

But it's like this, those guys that I stuck at #1 win about half of their matches and they are really are 3.5 players, if you think they are too weak and you are not having fun, then go play in the 4.0 league if all you care about is "being competitive".

In reality I think there is a difference between "order of strength" meaning the difference between each positions chance of winning, versus whether you get a good match or not.

It's not supposed to be like #3 is a bunch of 2.5 players and #1 is 4.0. It shouldnt make THAT much difference.

(usually when I identify a position that's potentially strong a lot of times it doesnt always mean they are over-skilled, sometimes it just means they make a good partnership and they seem to have a good win rate which could be attitude or just about anything)

Jim A
02-27-2009, 02:53 PM
so a lot of teams rotate into different slots each week, keep the pairings random the 1/2/3 doubles or 1/2 singles?

Is stacking putting your better players up against supposedly weaker players?

Never even gave that stuff thought as I head into my 1st NTRP season

JavierLW
02-27-2009, 02:56 PM
Is stacking putting your better players up against supposedly weaker players?



Yes or it can be putting your better players up against their better players if you suspect they are stacking and you dont want to have a mismatch.

Or you think that they think that you are stacking a certain way so you stack a whole different way to get a result that is different from what you are doing. (I did that to one team because I knew their captain actually looked at that sort of thing so I tryed to guess what he was thinking)

As a player, it's best not to even get involved in that sort of thing. Just show up where you assigned, and play tennis and have fun. :-)

No matter what number you are playing it's likely that if you play your best you can win, and if you dont play well you can lose to anyone.