(Seemingly) Irrational alliance selection decisions?

[TDav540]

Originally Posted by SoftwareBug2.0

I also took a whack at reviewing alliance selection results from this year. The way I did it was to say that a pick made sense if there was some way in which to rank teams in which the one that they chose was in the top five of those still available. I found a ton of picks that I just could not account for.

I thought I'd see a couple alliances per event that was just doing pure OPR based picking or pure rank, but it turned out that for the events that I was looking at those weren't the case at all. And a lot of them didn't make any more sense after looking at scouting data. But I did find that there was a positive correlation between alliances that made picks that I could figure out and advancing further in the tournament.

It seemed that many captains picked so badly that they would have increased their odds of winning the tournament if they had just been picking based on rank. I can only conclude that are large number of picks are often made for reasons other than maximizing the odds of winning the tournament.

And I think there are sometimes good reasons for trying to do something other maximize odds of winning, like the one that jajabinx124 cited with the wildcard system. And similarly to the wildcard system, in the district system how far you advance in the tournament matters, not just winning or losing the event.

I'm not sure how common this sort of strategy is however. For example, during the PNW championship I knew exactly how many points my team had to get to give us a 100% chance of making worlds, but talking with some other teams in attendance it seemed like a lot of folks didn't really grasp the point system well enough to figure out how well they had to do to get invited to worlds, let alone what that would imply strategically.

One extra note I will give is that it was easier to predict selections as the level of play increased. For example, while some in the audience and even the MC found the alliance selection on Hopper surprising it really shouldn't have been a surprise to any of those involved.

It all depends on your goals. Some teams want to make semis, some finals, some win. Some just want to make the playoffs and have a good time.

Additionally, although it may seem to us (especially in hindsight) that a team's picks didn't make sense, they may have a completely valid reason. Some teams have untapped potential (I've worked with a few of those cases), others have (or at least think they have) good chemistry, others still have complimentary strategies. It may be an incorrect assumption or decision, but it's what that team believed at the time.

How often do teams make the right pick? There are enough variables to consider that really teams should be just looking to make their best possible pick. At the end of the day, doesn't the "right" pick depend on the result?

[Chris is me]

I'm not really a fan of the word "irrational" to describe picking behavior. There is too much subjectivity in picking and information for all participants for this to be the term I'd prefer. Most every alliance selection pick is maximizing the value of their pick as best they can, which is by definition rational. Many examples brought up in this thread are easily defendable picks (like 973 picking the strongest scorer in their division, or a team declining for a better chance at a wildcard).

The truly terrible picks are from teams that are unprepared, don't care enough, or don't have the ability to get the data they need to make a good pick. Usually a combination of factors here, as even teams without formal data can make decent picks sometimes, and even teams who don't care enough might pick a first pick with good information to work with.

The odd, suboptimal but workable picks happen for a variety of reasons. Different seeds have completely different incentives for their alliance picks. The number 1 seed at regionals / districts generally wants to eliminate variance as much as possible, picking a consistently strong first pick and a consistently functional second pick. The 2-4 seeds want teams that may be less consistent but have a high enough ceiling that with some luck they can outscore the top alliance. The bottom half of the draft either wants to go for a strategy where 3 decent scorers beats 2 great scorers, or they want to pick super high variance teams with high point ceilings and hope for some luck and preparation to help them through the eliminations.

I'll add more thoughts later if I get the time.

[Christopher149]

Originally Posted by TDav540

It all depends on your goals. Some teams want to make semis, some finals, some win. Some just want to make the playoffs and have a good time.

Some are at district champs and want to help certain teams get to CMP.

[Jon Stratis]

I wouldn't say any alliance selection decisions are irrational. Some are unprepared, some are based on faulty data, and some are based on different value propositions.

Perhaps the biggest thing to remember during alliance selection is that no one has perfect information. Every team representative down there is dealing with a state of asymmetrical information, in which every team is basing their decisions on different information. Sure, you can say that they all have access to the same information - match results, for example - but the process of collecting data and processing it into a usable pick list causes the data itself to undergo a different transformation for every team, and end up being viewed differently by every alliance representative. As a result, we end up with picks that may or may not agree with any given set of data or viewpoint.

Honestly, I find the uncertainty going into alliance selection rather exciting.

[TheBoulderite]

Another seemingly irrational alliance selection this year was when 1425 selected 1538 on Hopper. Don't get me wrong, 1538 is a good team, but they could have gone with a team like 971, 1323, 4334, or 4587. Anyone from 1425 want to chime in and tell me why you made the choice to go with the Cows? I'm not saying it's the wrong choice, I'm just curious.

[Koko Ed]

The craziest situation I ever witnessed for alliance selection was at the 2009 FLR where 610 was a 13th seed and turned down an invitation from 1765 who had their best weekend in their history that weekend. 610 was taking a huge gamble that the other alliances were going to pick one another and they'd settle in as an 8th seed but for that to happen they would have to hope that 340 would pick their little sister team 424 which they didn't and thus 610's day was done.

[OccamzRazor]

Hey did anyone watch Curie division picks? 100% upset brackets this year in 2016. Every alliance selection made me cringe until the 7th or 8th seed. (which hilariously enough were division finals)

[TheBoulderite]

Originally Posted by Koko Ed

The craziest situation I ever witnessed for alliance selection was at the 2009 FLR where 610 was a 13th seed and turned down an invitation from 1765 who had their best weekend in their history that weekend. 610 was taking a huge gamble that the other alliances were going to pick one another and they'd settle in as an 8th seed but for that to happen they would have to hope that 340 would pick their little sister team 424 which they didn't and thus 610's day was done.

Yikes. That's crazy.

[TheBoulderite]

Originally Posted by OccamzRazor

Hey did anyone watch Curie division picks? 100% upset brackets this year in 2016. Every alliance selection made me cringe until the 7th or 8th seed. (which hilariously enough were division finals)

Curie was an anomaly this year. Some of the better-known teams like 3310, 1983, and 2848 weren't as strong at Curie as I thought they would be. However, they were good enough to be selected for eliminations, and the captains thought they could win with them. I didn't cringe, I was curious to hear the strategy behind making those selections.

[NShep98]

The only pick I have personally seen that would come anywhere near "irrational" for me was when one alliance picked the lowest seeded team, then the alliance immediately after that picked a much higher seeded team, where the two teams were the same 4 numbers mixed around. It may have been confusion, it may have been a valid pick, I don't know, but it seemed off to me.

Originally Posted by Zac

One of my favorite alliances to be on from this past year was at the Boston regional. 1768 was with the alliance captain 1058, as well as 5563 (they actually came in as a backup robot early in the eliminations, and were an absolute jackpot). I have never felt so comfortable on an alliance. 1058 did what the were doing best and made sure to get boulders in the low goal for the guaranteed tower strength, this simultaneously gave the rest of the alliance a huge sense of relief. All of a sudden there was so much less stress to score X number of boulders, 1058 has that covered. This meant less stress for the other drivers. 5563 was absolutely awesome, and knocked down the defenses while being simultaneously invisible most of the time, they were cautious in when they crossed as to not disrupt any shots that were being taken, and they always stayed close to the defenses to allow for easy driving around for the rest of the alliance. 5563 also allowed for 1058 and 1768 to climb on the outside positions of the tower, and would then drive up the batter at the last second to avoid bumping either robot during the critical "get the hooks on" phase. It was fantastic. Each robot on the alliance allowed the others to play to their strengths.
~Zac

Meanwhile, on the other side of the glass from this, we were the backup team for 5422 with 125 and 238. After making it into the finals (marking a District event where both finalist alliances had 4 teams, no less), we were told before one of the matches that scouting teams had had their eye on us, and the reason no one was told to pick us in the first place was because they forgot. No disrespect to the alliance captain or the team we ended up replacing, but not a fun position to be in.

[Maximillian]

Originally Posted by TheBoulderite

Another seemingly irrational alliance selection this year was when 1425 selected 1538 on Hopper. Don't get me wrong, 1538 is a good team, but they could have gone with a team like 971, 1323, 4334, or 4587. Anyone from 1425 want to chime in and tell me why you made the choice to go with the Cows? I'm not saying it's the wrong choice, I'm just curious.

1425 selected 971 initially and 971 declined.

Originally Posted by AdamHeard

in SF 2, all 3 teams missed in auto and performed on the lower end of our averages. This combined with a great performance from 2122, 3538 and 2052 cost us the bracket.

This sounds like an all too familiar situation.... It's a shame when all your auto's decide to miss.

[TheBoulderite]

Originally Posted by Maximillian

1425 selected 971 initially and 971 declined.

I was unaware of that. Thanks.

[frcguy]

Originally Posted by TheBoulderite

Curie was an anomaly this year. Some of the better-known teams like 3310, 1983, and 2848 weren't as strong at Curie as I thought they would be. However, they were good enough to be selected for eliminations, and the captains thought they could win with them. I didn't cringe, I was curious to hear the strategy behind making those selections.

Disclaimer: My team was the 4th pick for Curie alliance 4, which included 5803, 3310, 2168, and us (5940)

I don't think that 3310 was a irrational pick for our alliance. Yes, they hadn't performed great in quals, but if they were able to put in 8-9 high goals as they did at the regional level it would have been huge for us and would have, in my opinion, changed the outcome of eliminations in Curie. 3310's prowess in the high goal + 2168's ball-hoarding strategy from NE DCMP + 5803 breaching would have been an incredible combination. 3310 had everything working great going into the quarterfinal matches. Unfortunately, they suffered some unpredictable failures (such as a busted VRM in QF2-2) that decreased how many shots the could get in.

In short, if 3310 was working at full capacity I believe Curie might have been a different story.

[D.Allred]

Originally Posted by BMSOTM

My questions for everyone:

• Why do teams behave (seemingly, to an outsider) irrationally during alliance selection?
• Has anyone noticed any patterns or common fallacies*?
• How frequently do teams make the "right pick" or something close to it? Is there a way to quantify/measure this?

*Aside from a lack of scouting data. I know from first-hand experience that it can cause problems for an alliance captain.

Why do teams behave irrationally?
Like others have stated, there might be a rational reason that doesn't align with your pick preferences. Scouting data alone or OPR results won't necessarily build a good alliance. Alliance "chemistry" and relationships are important as well.

Has anyone noticed any patterns or common fallacies (other than scouting)?
I have no data to support this, but I assume the average FRC team does not have an elimination alliance plan going into the regional. It's difficult to scout if you don't know what you want. Even if you do have an elimination strategy, it is not easy to form a balanced alliance to fit your desired strategy or adjust to evolving game play.

This gets more complicated when you suspect a team would be better off as a role player, but continue to pursue their original game plan. Objective scouting data doesn't help you make a decision if they would be a good fit. Sometimes it works, sometimes not.

How frequently do teams make the "right pick" or something close to it? Is there a way to quantify/measure this?
I don't know of a statistical way to measure it, nor do I believe it would have meaning. To improve our scouting, we usually compare our pick list to actual alliances to see if we missed something in our scouting results. There have been several surprise "right picks." We usually rely on a few objective data points plus subjective comments to make a pick list. Sometimes the more subjective elements were key but not understood at the time.

David

[SpaceBiz]

I think there is a significant difference that needs to be clarified between irrational picks and bad picks. Team's make bad picks all of the time, but There was always at least some rationale behind any pick. That selection had to arrive on a picklist for some reason, whether or not it was a good reason. If a pick seems irrational, there might be a good reason behind it. If you have outstanding scouting you can find value in teams they themselves might not even know they had, let alone the general public.

I think the more important thing to look at was the result the pick achieved you and if they yielded the result you expected them to yield. Essentially was the pick a bad pick. If you select a high variance team with high scoring potential and it doesn't come through, but you were aware of that risk beforehand, the pick is not particularly bad despite possibly yielding a bad result. If you select one team for some reason while leaving another more valuable team on the table, and this selection ends up hurting your performance, it can be argued that you have made a bad pick.

Defining whether or not you think your pick was bad or not is important. Once you do, you can then look at the reasoning you had to pick that team and then find what flaws you may have had in that reasoning to yield the wrong result. Hopefully, you will be less likely to make that mistake in the future.