What Is Incentivized Due To Double Eliminations?

I’d like to start a public discussion on what characteristics of robots or teams are incentivized or downgraded due to the new structure of the playoffs. For instance, the fast turnaround times of past tournaments might have made robots that were very simple (and thus less likely to break) a higher second pick, while the new structure guarantees a certain amount of time between rounds, perhaps allowing picking teams to choose higher risk/higher reward robots for their second pick. I wonder if the community has any insights or speculation on this. It will likely influence teams in their early design and strategy decisions.

11 Likes

We will definitely keep reliability in mind even more so than in the past. At all of our offseason events, at least one alliance robot broke during playoffs. Those quick turnarounds are intense. Not breaking/being fixable will be very important.

14 Likes

The classic “if you’re a lower seed you gotta take risks” is less true than before - a low seed doesn’t have to go “upset or bust” since they can advance in the elimination bracket. I imagine I’d focus less on upside if picking from that spot.

20 Likes

I don’t think we’ll see much change at the top end of the draft. The top few alliances will pick the other strongest robots at the events, good driving, no breakdowns, etc.

I think where it will get different is the bottom of the draft. The #8 Alliance no longer needs a Hail Mary to advance in the tournament. Instead of #8 picking the robot that has a traversal climb for example but only did it once or twice because they need that miracle in order to advance might NOT get picked. The #8 Alliance may instead pick a more consistent, well rounded Alliance that may lose the first match but might be more consistent and better suited later on.

This is to say I think this format incentivizes robots that are consistent from the top of the draft to the bottom.

17 Likes

In the 3 competitions we’ve won in recent years, they each took us 8 of the possible 9 games to do it (that means we lost 2 of those, something that you can no longer afford). I think one of our strengths has been learning and strategizing from that that initial loss to win the next two.

That’s not going to be a viable plan going forward, you’re going to need to “hit the ground running” so to speak. I’m not sure a “maybe awesome/maybe not” robot pick is compatible with that. YMMV.

7 Likes

Reliability/consistency will be a bigger criteria than “peak” performance. The consequence of a breakdown, disconnect, etc will be higher. You can’t afford to have any bad matches.

I see this as the biggest difference. Alliances are going to have to figure out a winning strategy for each and every match. You won’t play against the same alliance more than once the way you did with the Bo3 tournament style so you won’t be able to adjust your strategy after the first match to better match up against the opposing alliance.

My takeaway from this will be that there will be a huge benefit to scouting the playoff matches to see the strategy that your potential future competition has employed so that you can put together a strategy that can match up well to it.

I also see teams taking less risks with their strategies. For example, in 2019, you could decide whether to pre-install hatch panels on the cargo ship. The lower risk strategy was to install all of the ones available to pre-install so that you could proceed directly to loading cargo (which was easier to do while being defended). The downside to this strategy is that it capped your max score at a lower level. So a higher risk strategy was to leave a few panels off so that you had the possibility of a higher score. The risk was that if you were not able to get hatch panels installed because the opposing alliance was defending you then you could also not score cargo in those slots. I see teams taking the lower risk strategy in the playoffs under the DE system.

9 Likes

Incentives will be based on team/alliance goals. Assuming the goal for a team/alliance is to advance to the world championship, we might see things play out differently in the regional and district systems.

In regionals, you still need to win the event to guarantee a spot. I believe that means that the 5-8 alliances will have exactly the same incentive structure as they do in the old format. They still need to beat the top alliances, and therefore still need to make high risk/high reward type picks.

In districts, the goal is just to accumulate as many points as possible. I could see 5-8 alliances prioritizing reliability over potential if they view their second/third round matchups as being winnable.

In regionals with tons of wildcards, I could see teams make picks that maximize their odds of getting to the finals, rather than their odds of winning the event. Granted those two things are usually going to be the same.

Ultimately I don’t think the incentives change very much, and likely won’t be perceptible to very many people outside of the alliance selection decision makers.

9 Likes

Something you occasionally saw at very competitive regionals was alliance captains declining in order to stay in the 3rd or 6th spot to try and angle for a wildcard. I think you will see that far less often now.

I also think you will see lower seeded alliances focus just a bit more on overall reliability and consistentcy, just a bit less on peak performance. But if the goal is winning vs just advancing as far as you can, the incentives aren’t that different - lower seed teams will still want to pick higher-variance teams over consistently mediocre teams, simply because it is the best odds they have of winning. It’ll just be a bit less likely to work out.

11 Likes

Outside the picking order, I think DE heavily benefits teams that have invested in scouting and strategy. Being able to come in with an unexpected strategy that catches a team off guard is way more viable without needing to win a Bo3.

In the same vein, I predict that we will see a lot more “off-meta” strats in playoffs, looking to use the shock factor in order to secure a win before the other alliance can adapt. (e.g. 2 Defense in 2022 would have been more prominent)

Something that comes up a lot is the power of a “gimmick” in Best of One formats. I’m interested to see if certain teams will try to hide certain functionalities until near the end of the tournament in order to utilize that surprise factor (which of course risks not getting picked because you didn’t show your full potential). However, I still think that the risk of anything like this is probably too great to even consider for most teams.

9 Likes

I was actually thinking of this, “gimmicks” or “one-and-dones” were highly effective during turning points, at least in eSports, as teams couldn’t react quickly to something that they’ve never seen before. In a BO1 format, every match is a turning point which ensures that a team with good strategy could use something that newly-formed teams have no immediate counter-strategy planned to shock them. I see this being incredibly impactful in the first bracket match where teams aren’t used to playing together could be out-of-commission if something like ball control into hangar was played.

3 Likes

Honestly, because of the nature of one-off matches and tight turnarounds, I wonder if there will be a resurgence of teams attempting to find a way to push for cheesecaking again. Having an unscouted pocket strategy or mechanism would be such a wildcard to throw into this format.

5 Likes

This also might combine with backup robots being used to switch up an alliance’s playstyle from match to match.

Definitely something to watch out for more and more this season.

Did indeed forget about that rule change, I don’t see this being used much unless an alliance had a pure defense bot that could push around a bot but I still see three offense bots with a 2 offense 1 flex strategy being the main default, unless this year’s game is heavily different from previous games.

1 Like

How late can you decide your line up? Can you have all 4 robots in queue and try to hang back and see which 3 robots your opponent fields? Can you fake your opponent by walking a decoy bot in the in gate and then right out the out gate after they’ve noticed it, and then place your actual third? etc etc.

1 Like

This is also a good point, I hope FIRST does something about realtime lineup changes as in having alliances submit a lineup right before the match, hopefully match strategy doesn’t devolve into sleight of hand and last minute monkeying.

1 Like

I agree 100%. Not having to Hail Mary with a risky strategy to knock out your initial red-bumpered opponent who, let’s face it, almost definitely outclasses you (at least at the regional level) should inform the 5-8th seed picking strategies.

The “win vs this one opponent or go home” is no longer there, so you don’t have to plan for trying to upset the 1st seed 2 out of 3 times.

Assuming it works the way submitting lineups worked at champs for years here is the relevant rule 11.9.2 from last years manual

ALLIANCES may start with any 3 of the 4 ROBOTS on their ALLIANCE during division Playoff MATCHES and during the Championship Playoffs. The list of 3 teams participating in the MATCH and their selected DRIVER STATIONS is called the LINEUP. A single representative from the team not on the LINEUP is allowed as a sixteenth ALLIANCE member. This additional representative may only serve in an advisory role and will be considered a COACH (e.g. can’t be a HUMAN PLAYER).

The LINEUP is kept confidential until the FIELD is set for the MATCH, at which point each ALLIANCE’S LINEUP appears on the Team Signs.

If an ALLIANCE would like to change their LINEUP after their first division Playoff or Championship
Playoff MATCH, the ALLIANCE CAPTAIN must report the LINEUP to the Head REFEREE, or their designee, in writing prior to end of the preceding MATCH (e.g. the LINEUPS for Quarterfinal 2 must be submitted before the end of Quarterfinal 1). If the Head REFEREE is busy and there is no designee defined, the ALLIANCE CAPTAIN waits in the question box to report the LINEUP.

Once the LINEUP is declared, it cannot be changed unless there is a team or FIELD TIMEOUT. If there is a TIMEOUT, the ALLIANCE CAPTAIN may submit a different LINEUP, but must do so while there are still more than 2 minutes remaining in the TIMEOUT.

3 Likes

So you could queue up robots A, B, & C, whilst your captain slides on over the question box [head ref is going to be busy, that’s a given] with one or more lineups (possibly some including robot D) in hand. This is getting a major spycraft vibe :wink:

4 Likes

I dislike the timeout switching since it is a BO1, this is sort of what I mean though.

1 Like

Well there is no more time out coupon so the timeout only applies to field timeouts. Imagine you are the winner of the top half of the bracket, if this timeout rule is not in place then you have to submit your lineup before you know who you are facing in the finals. It is even worse for the 2 alliances in the lower bracket final since they would have to submit their lineup before their match starts not knowing if they are getting out or not or what state anyone is in.

This same logic can apply to anyone who has to play back to back so also includes the teams who are heading into the lower bracket final since they both would have to submit a lineup under the same unknowns.