Strategy during Final at Einstein

[Hoover]

We were watching Einstein live from home. The first match spread was 8 point and the second match's was 10 points. Very exciting and good quality matches by all teams.

After the second match, my daughter and I agreed that 1690-2056-1405 should change from a defensive game to a strategy where all three teams go on the offensive.

The reason for this is math. For 1690-2056-1405 with two robots scoring and one on defense, the scores were roughly equal to 120-2481-330 with their three robot offensive. Say the defensive strategy impaired 120-2481-330 to strength of 2 1/2 offensive power. Then doesn't it make sense 1690-2056-1405 were good enough with two scoring robots, that a third robot on offense could have taken the third match?

Additionally, a defensive strategy was where downside risk of fouling dwelled this year.

[EricLeifermann]

1405s defense is why the scores were close. With out the defense the 330 2481 120 alliance would have pulled very far ahead.

[rick.oliver]

Their strategy was correct, in my opinion. In fact, after their first Quarter Final match, I thought that they should switch to a defensive strategy ... and they did.

It served them well to get to the Final matches.

[GKrotkov]

I agree with the above two posts - defense wasn't their issue, it was a huge benefit to their alliance. In fact, I'm surprised that more Einstein alliances didn't play defense like Tesla's alliance did with 1405 (direct defense), or like Hopper's alliance did with 2990 (defense through boulder management).

[Chris is me]

1690 and 2056 could score quite a lot of balls on their own. If it were 2v2 versus 330 and 2481, I think 2056 would have won. 1405's defense was less good against 330 and 2481 because 2481 could line up and shoot quickly and 330 was a tall batter shooter, but it was still their best chance and undefended the Carver alliance surely would have run up the score. 120's offense and hanging made a huge difference.

[g_sawchuk]

Their defense was, without a doubt, necessary. Without it, the scores would've been separated by a larger margin. 1405 typically stopped quite a few shots.

However, one thing to note is how vulnerable 1405 was to defense, and how easily Einstein Finals could've been won in 2 matches.

1405 typically made some close calls in regards to when they left to capture. This was clearly evident in Einstein Finals 1, where the blue alliance failed to capture due to 1405 not making it to the batter. In Einstein Finals 2, 1405 was able to get to the batter by 8 seconds, leaving at just under 30 seconds. The red alliance, at 30 seconds, could've simply flipped the courtyard roles, playing defense on 1405. They would aggressively prevent them from reaching the SP for as long as possible. The result? Pretty scary for 1405. They can't leave through a defense, because the red alliance would be close by, and contact would result in the blue alliance being fouled. You could extend this defending of 1405 until past 20 seconds, but as this point you could raise concerns/debate in regards to if you're trying to draw a foul (of course, you're not, you're just trying to prevent a capture).

All in all, if you held them until 20 seconds, based on how long it typically took them to make it securely to the batter, the red alliance would have achieved a capture, and the blue alliance would not have.

Just like that, Einstein Finals would be over in 2 matches.

[barn34]

I can assure you that they made the right call going defense against us. In fact, I was hoping they would have made a change to all offense because I knew that was a favorable match up for us.

With 120 in our lineup, we averaged 229 points.

In matches our alliance was left undefended, we averaged 243 points.

On Einstein, when facing defense, we averaged 215 points.

Tesla averaged 221 points with 3015 in their lineup (more offense)
- if you remove their worst match outlier of 165.

Tesla averaged 213 points with 1405 in their lineup (heavy courtyard defense)
- that's only a delta of 8 points of offense sacrificed with the potential of reducing our potential offensive output by 28 points down to 215.

213 vs. 215

or

221 vs. 243

Which situation would you want?

Going with 1405 on defense put them essentially even with us on the potential scoreboard, and that was represented in how close the final matches played out. These matches were essentially a coin flip in this scenario. Letting us run wild on offense would have been a much more difficult scenario to overcome.

With that said, we did just enough in the midfield to slow down 2056 so they couldn't get that 1 more ball that would have made the difference. Any time we saw them in the midfield, we looked to make them work for it. Anything to cost them an additional second or two. Over the course of a match, that can add up to potentially be the difference when it's so close. Turns out, that's exactly what happened and 1 foul was the difference.

I've said it before and I'll say it again, we're both champions in my book. #2champs a year early.

[BenHildy]

I would agree wholeheartedly that the defense played by 1405 was necessary in the finals against a team with as much offensive power as the Carver winners. My reasoning is this:

1) The effective use of defense more or less shut down 120 in their attempts to shoot high goals. Furthermore, 330 preferred shooting from or near the batter, which is much easier (and safer) to defend against than 2481's consistent outerworks shot, but is deadly if left alone.

2) The offensive capability of OP and Orbit was certainly enough to go shot-for-shot with the opposing three robots while they were inhibited by defense, so the alliance wasn't losing much offensive power to a member playing defense.

3) The most important thing to do against that Carver alliance was to slow them down, which was done well by 1405. When left alone, I noticed that 2481 was winning nearly every neutral zone "Boulder battle" that they entered, and Carver could've quickly beat out Tesla using a choke strategy. However, since 330 was taking a few extra seconds to shoot every time they cycled, this gave OP and Orbit the time they needed to grab the balls before 330 and 2481 could get to them.

Overall, while defensive penalties proved to be their eventual downfall, I believe that the strategy Tesla took was certainly appropriate for the opposing one. Frankly, I was just relieved to see defense being played on Einstein at all, because I noticed that many of the alliances ended up being three high-goal shooting robots. No offense to these alliances by the way, I just prefer to see a varied strategy on an alliance. Good job to all and special regards to the Carver Division Alliance!

[bam-bam]

If 3015 and 1405 didn't play defense, they would have likely lost to Archimedes... in quarterfinals. Defense was definitely the right call, and I'm very impressed that 2056 had the foresight to pick up 1405.

[Brian Maher]

Defense was definitely the right call, and I'm very impressed that 2056 had the foresight to pick up 1405.

I would hypothesize that 3015 had something to do with it. Both 3015 and 1405 competed at the Finger Lakes Regional, where 1405's defense skill helped their #5 alliance with 20 and 639 upset the #4 alliance, the #1 alliance, and the #3 alliance. The #1 and #3 alliances definitely could outscore #5, but 1405 gave them the defensive edge to take home banners. 1405 plays superb defense, and the Tesla #1 alliance was lucky to be able to pick them up.

[BigJ]

I would hypothesize that 3015 had something to do with it. Both 3015 and 1405 competed at the Finger Lakes Regional, where 1405's defense skill helped their #5 alliance with 20 and 639 upset the #4 alliance, the #1 alliance, and the #3 alliance. The #1 and #3 alliances definitely could outscore #5, but 1405 gave them the defensive edge to take home banners. 1405 plays superb defense, and the Tesla #1 alliance was lucky to be able to pick them up.

2056/1690 actually announced both picks simultaneously, so there wasn't any lobbying from 3015 (as far as I know). In fact, they were some of the fastest decisions in the Tesla selections overall :p

[throwaway]

I was surprised that 330-2481-120-1086 didn't employ much defense against 2056-1690. OP in particular was really vulnerable as they were doing mid-court shots that they could not have gotten away with under heavy defense. Had their been defense played against OP I think their ability to score would've been dramatically inhibited and I doubt the finals would've gone to 3 matches.

[wilsonmw04]

I was surprised that 330-2481-120-1086 didn't employ much defense against 2056-1690. OP in particular was really vulnerable as they were doing mid-court shots that they could not have gotten away with under heavy defense. Had their been defense played against OP I think their ability to score would've been dramatically inhibited and I doubt the finals would've gone to 3 matches.

There were discussions, but it's hard to argue with the results from Saturday.

[TDav540]

I was surprised that 330-2481-120-1086 didn't employ much defense against 2056-1690. OP in particular was really vulnerable as they were doing mid-court shots that they could not have gotten away with under heavy defense. Had their been defense played against OP I think their ability to score would've been dramatically inhibited and I doubt the finals would've gone to 3 matches.

As we've seen before, OP can basically shoot from anywhere. Their percentage may have decreased, but that's no guarantee, especially over such a short sample size. In that situation, it was probably better to let both 120 and 1086 play offense, which is a more comfortable role for them based on their prior experience.

There were discussions, but it's hard to argue with the results from Saturday.

This pretty much sums it up.

[FIXIT]

Team 1405 played awesome defense and really slowed us down. It could have been anyone's victory in the end. It was one of the greatest battles of all time. Fortunately for us we won! It could have gone either way. Both alliances were fantastic and did not disappoint the crowd.

Congratulations to everyone for a hard fought and memorable showing on the final Unified Einstein World Championship. What a great way to end an era!

[barn34]

I was surprised that 330-2481-120-1086 didn't employ much defense against 2056-1690. OP in particular was really vulnerable as they were doing mid-court shots that they could not have gotten away with under heavy defense. Had their been defense played against OP I think their ability to score would've been dramatically inhibited and I doubt the finals would've gone to 3 matches.

That's where OP was shooting while undefended, but they were very capable shooters from the outerworks, as well. It would have neutered a chunk of our offensive output and probably not impacted 2056 much, if at all. They would have just backed up and shot protected with their intake out to create space and probably draw some fouls. Plus, 120's robot wasn't particularly built for playing defense. Their drivers also let us know that they didn't think they'd be as effective at it. If we were going to go that route, we would have needed to sub in 1086. That meant we would have also taken the climb points off the table in the endgame which was really helping us stay afloat with 330's climber issues lingering following SF2. 1086 also let us know earlier that the new Einstein balls may mess with the accuracy of their shot, which they had tuned using worn down balls. That proved to be true based on the accuracy seen in SF1 when they were on the field with us while 120's intake was being frantically repaired. We would have needed that autonomous shot to hit to offset the lost scale points. Even if 120 was getting shut down, that was still pulling a lot of the defensive attention away from us and 330.

Defending OP had to be done in the neutral zone. In match 1, we got a couple good blocking and slow down plays on them as we were chasing down boulders. In match 2, they did a great job avoiding those situations. In match 3, we again managed to find them a couple times and did just enough to be a nuisance.

I just want to point out that match 1 could have very easily gone the other way if either of 2 things happened in the endgame. If 1405 was able to get on the batter that would have been an additional 30 points and the score margin widens significantly. If OP would have gotten the scale, that would have been 10 more points and they win by 2. If both happen, that's a 40 point swing and we lose by more than a capture bonus. In match 2, they basically got us by an auto ball. Match 3 was essentially a draw. These finals were crazy close, obviously. Completely swung on just a few plays.

[throwaway]

As we've seen before, OP can basically shoot from anywhere. Their percentage may have decreased, but that's no guarantee, especially over such a short sample size. In that situation, it was probably better to let both 120 and 1086 play offense, which is a more comfortable role for them based on their prior experience.

That's not true. OP essentially lost their ability to do outerworks shots after GTR East, if you watch Waterloo, Windsor, and Tesla that's very apparent. And while obviously the 330-2481-120 won my argument is that it would've been a much easier victory for them if they had use defense against OP.

[wjordan]

I am still confused as to why 1690 went undefended for the majority of Einstein. In the matches in which they faced defense at Tesla, their offensive output dropped tremendously.

[Rangel(kf7fdb)]

I actually do believe defense was the right way to go for Tesla alliance. However, I do agree that overall, triple offense was the bread and butter for most Einstein teams. For one, it's just a lot more consistent. Sure the scores can vary by a couple boulders but when you dedicated one of your robots to defense, that defensive robot regardless of how high of skill they are, can go from shutdown defense to not very effective at all. I think for the sake of consistency and confidence in scoring is why most Einstein matches were shoot outs. It's sort of why I think the game this year falls a little short at the highest level. That being said, I think Tesla realized throughout the course of Einstein matches though, that they were unfavored in a shootout and switched it up.

[wilsonmw04]

snip

That's about right. The loss of accuracy was based on me forcing my drivers into shooting at an angle they hadn't practiced much. In my mind, I didn't want to clog up the middle of the field where 330 and 2481 were shooting. If I had to do it over again, I would have had us go under the low bar and park in defense 2 or 4 for the shots. We did lose a bit of length from our shot with new balls, however, length wasn't our issue. We had the length, but not enough practice finding the right spot and angle on the side goals.

I know a coach and a few drivers that would love to do that match over again. Live and learn, right?

[Citrus Dad]

2056/1690 actually announced both picks simultaneously, so there wasn't any lobbying from 3015 (as far as I know). In fact, they were some of the fastest decisions in the Tesla selections overall :p

Not uncommon on the #1 alliance. In Hopper we had 364 and 2990 as side by side picks on our draft list so it was easy.

[Electronica1]

The 3 team offence was the right strategy in my opinion since it minimized the risk of penalties. The only change I would have made was to put a taller defense near the low bar. The blue alliance's human player was able to get a decent number of boulders all the way to the opposite side of the field in some of the matches by going over the ramparts.

[LeelandS]

The 3 team offence was the right strategy in my opinion since it minimized the risk of penalties.

This was actually the reason our team felt so comfortable playing defense. Oliver, our driver, is an exceptionally smart driver, with a great skill for foresight and controlling the field as he wants. It was his knowledge of the rulebook and awareness that kept our penalties to a minimum.

At Finger Lakes, where we were our alliances defensive robot, we drew 2 penalties in the quarterfinals, and then we did not draw an additional penalty afterwards. And those penalty-less matches were pitting us against the likes of 5254, 2791, 3044, 4930 and 340.

In playoff matches at champs where our robot was fielded, our alliance never drew more than 1-2 penalties. In any match where we did draw penalties, the opposing alliance did not successfully capture the tower, so I believe one might argue that our defense contributed more to our alliance than against it. The only matches where this streak falls through, ironically, are the final Einstein matches.

The penalty we drew whilst in the 3rd final match on Einstein was hard to avoid, as Oliver had no visibility of the courtyard referee (a recurring problem for him all throughout the event). As many have noted, it was a tough call by the referee to begin with, coupled with the fact he was having information relayed to him instead of seeing it with his own eyes, it would have been quite a feat for him to have been aware of the whole situation. But hey, that's the risk we took.

In the end, it couldn't have been any closer. As far as I'm concerned, all 8 teams deserve to be world champions. It just so happens that the record books with not have our alliances names in them. It was a disappointing loss, especially given the fact that it was such a crazy alliance to work with (3 different countries represented, and allied with another team from our region). It would have been a good story to tell, but nevertheless. Congrats to the champs, they were hard fought matches and I'm happy with our season in the end.

Edit:
In response to our choice of strategy, 3015 was constantly used throughout eliminations. In our first QF match, our line-up was 2056-1690-3015, and our alliance ended up putting in 250pts. It was quite a spectacle, but we found that that match was a "perfect scenario" and difficult to repeat. Essentially, 3015 with their superior shooting (to us at least) and being a good defense robot as well, was our advance team, and was fielded in the first match of each round. If it was found that we couldn't brute force our way against an alliance, then we took the field. We knew that if it came down to 2v2, 2056 and 1690 could overpower almost any other duo, so all we had to do was take our one robots worth of shots, and that sealed the deal all the way to the finals.

[TDav540]

That's not true. OP essentially lost their ability to do outerworks shots after GTR East, if you watch Waterloo, Windsor, and Tesla that's very apparent. And while obviously the 330-2481-120 won my argument is that it would've been a much easier victory for them if they had use defense against OP.

Regardless, you're still asking 1086 and 120 to come out of their comfort zone and play a defensive role. While I'm not saying they couldn't do that, they definitely weren't used to it. In addition, while OP's outerworks shots definitely weren't the same, they still featured an accurate batter shot.

While we will probably still have our own opinions, if defense was going to be played on 2056 and 1690, it was more worthwhile to do so in the neutral zone, as previously mentioned. At least in my mind.

[Hoover]

Kudo to civilized banter; I thought it might be difficult, but this is reassuring. Nice input from teams involved so thanks. Even if we had been there we wouldn't have been able to view all the division brackets at once so there is not much back story of Einstein alliances.

Looking up some of the stats, it is hard to make the case that switching to an offensive lineup might have helped. The first thing I noted was that Carver's OPR's were significantly less than Tesla's going into Einstein. And no wonder, Carver's was the second seed alliance. (more on this later)

So from these numbers, it is harder to believe that Carver could keep up with Tesla. Therefore, defense had a significant impact on the matches.

Carver Seed 2 OPR
The Beach Bot 330 55.94
Roboteers 2481 50.96
Cleveland's Team 120 43.79
Blue Cheese 1086 43.07

Tesla Seed 1
OP Robotics 2056 68.08
Orbit 1690 69.8
Ranger Robotics 3015 36.48
Finney Falcons 1405 25.91

Speaking of OPR's it is notable that in most division the top two teams offensive rating was past the knee of the curve. Except for Newton but this has already been covered heavily on CD... but the notable thing with Newton is that almost all of those teams were upset even before Einstein. Then Newton was shut out of Einstein immediately.

However, it is more a question to my mind that more divisions are not like Newton was. Are high scoring robots on that end of the spectrum that rare, that in each division only 2 or 3 stand out even on a national level? I expect this in the districts, but nationally it is harder to rationalize.

[Chris is me]

Looking up some of the stats, it is hard to make the case that switching to an offensive lineup might have helped. The first thing I noted was that Carver's OPR's were significantly less than Tesla's going into Einstein. And no wonder, Carver's was the second seed alliance. (more on this later)

It's quite often a bad idea to try and compare OPR across events (divisions) as a measure of expected offensive performance of specific teams on the field. Perhaps Carver was a more even / deep field than Tesla, or Tesla had a stronger top end and a weaker bottom end, etc. Lots of reasons OPR is not a good way to quantify offensive performance in this context.