MSC Team Ranking Changes pre to post MSC 2014

Dear CD,

FRC3548 finally dialed in our 10 point Auton Shooter for MSC and scored 9 out of 11 times. In the Howell district we used a 1 pt auton. Even with that significant improvement in robot performance our pre MSC to post MSC ranking did not change (rank 60ish/64).

So, I ran a quick ranking evaluation of all the teams attending MSC and simply looked at the delta change pre and post MSC. I found some teams that had significant decreases in ranking and others that had significant increases and I was wondering why these changes occurred. Did your robot break down during MSC (107 dropped 49 in rank, 2054 minus 29) ? Did your drivers have a lot of practice post districts on a practice bot? What do you attribute your significant change in state ranking (2834 - plus 39) (910 - plus 44)? Do you think you had an “easier” or “harder” match schedule?

Are you willing to share your “secret sauce” so 3548 and others can improve for next year?

Team # MSC Rank Delta Team # MSC Rank Delta
Post - Pre Post - Pre
107 -49 4003 24
2054 -29 245 25
5193 -29 1506 26
3602 -28 3641 33
2851 -28 1918 36
3175 -28 2834 39
4391 -27 910 44
815 -25

There are many possible reasons. Here are a few that I can think of:

  1. Robot issues at an event. If you have some issues at a district but get them fixed you could expect to climb. If your issues were at MSC, you can expect to fall.

  2. Event strength. Teams that attend stacked districts are more likely to climb at MSC while teams that attend lesser districts are more likely to fall. Example: a team may do great at a district without much depth, but may only be the 10th best robot at a district like Waterford. That means the points that a team earns at a district depend on the depth of the district.

  3. Luck of the draw. Schedules do have an impact.

  4. Outside factors. Sometimes a great partner team’s robot breaks during a match. If that happens in the rubber match in elims, that can create a big qualifying point swing. Likewise for scoring issues, penalties, etc. Or, a key match in qualifications can juggle the draft order which could cause a big change in how points are awarded.

  5. Alliance needs. A great qualifying robot may not be what is needed by alliance captains causing them to slip down or out of the draft.

Marc,

I think most of the change in ranks came from the Alliance Selection Points.

There were teams that won events that didn’t get selected or got picked up late in the draft. Most of this comes from Chris’ statements about Alliance Needs.

-Clinton-

For 910, I think it was 3 factors…

  1. Robot improvements - Our first two events showed weaknesses in our winch and gatherer, and we didn’t play “smart.” We were semi-finalists in those two, but by the time we went to Buckeye, we had those fixed and got wins there and at Troy. We always look at what we can improve on and work like crazy between events to make ourselves better. We test it on our practice bot and implement it at the next competition.

  2. Driver practice - Every event they got better. They found just the right balance between offense and defense. They’re never sitting still. They also got better at playing while being heavily defended and also at shutting down opponents. They also learned to “tuck” the ball while driving to eliminate “pop outs” from hard hits by our opponents.

  3. Strategy / Teamwork - Watching our own “game films” along with hundreds of online ones from around the country helped us determine how to play better, and as a unified team. Having a plan ahead of time to switch plays on the fly based on if we have an opponent shutdown for example helped us keep scoring while preventing the opponents from doing so. Flexibility and teamwork often helped our alliance beat a stronger one that might not be playing as “one mind”.

We’re just hoping that we get the kind of awesome cooperation / teamwork at CMP that we have in Michigan!

Cheers!

Just something I wanted to point out - Our team only dropped 11 places, not 28.

And to answer your question, we had robot problems on Thursday with our auto not working, and once we got it working (sort of), our code overall stopped working, and we had electrical problems, a joystick die in one match (Q53) and also had a chain slip off one match (Q32). Whenever we were working, our alliance partners made us play defense, either because they did not trust us to get the job done, or they saw we had 6 CIMS on our drivetrain. There was two matches where we were allowed by our alliance partners to play offense. I honestly do not believe that our robot was the 64th best robot at MSC. Also, I believe that our schedule wasn’t one of the easiest ones at the event. Missing CMP by 3 MSC wins kind of hurt, especially when we went 2-10 :frowning:

I would say it was a combination of things. We had a moderatly tough schedule that by OPR stats suggested we would be 6-6 as we ended. you can see that here http://www.chiefdelphi.com/forums/showthread.php?t=128678&highlight=pre+event+data

The field was deeper than in the districts we attended although West Michigan had many of the same teams there and was quite deep as well.

I thought we played well but did not often show how our catching ability could be usefull inbounding from a human player.

I think we were percieved as not fitting the desired position needed by the alliance captains. We didn’t like it but that is how it goes sometimes.

I thought you looked very good. Your dedication and hard work showed.
Congrats

This is what I had heard a few times.

This was the case for the 1 alliance. We had 2054 high on the list of teams available, but when it came down to the pick, our strategy called for more of an inbounder/defender than a scorer/trusser.

When we played with 2054 earlier in the day, they did not feel they could “pass” the ball effectively without rocking the robot. We were concerned with their ability to inbound effectively and ball security.

We felt better with 74 being the inbounder than adjusting our planned strategy to fit for 2054.

Unfortunately, sometimes alliance selection comes down to the perception of which robot style best fits an alliances planned strategy. Making the right second pick can make or break an alliance.

I questioned why you picked 74 until I witnessed your strategy. They did a nice handoff to 1718 BOT to BOT which also kept the ball pretty secure. I believe we have an easier robot to pass into the when we go into catch mode and spread our “Wings” but we did not show that much. We also have a short range air cylinder “Puncher” that moves the ball back out to pass with but it would not be as seamless as the rotational movement of 74’s pickup running in reverse passing robot to robot with 1718. We could have inbounded and passed over the truss to the human but then you still have to get both other alliance members the ball for the assist and is a break from the strategy you had.

Not to mention we did not show these other capabilities much during seeding. In fact I don’t remember a case of us using our short pass cylinder that ejects the ball out our pickup mechanism side so who knew we had it or what it could do? I don’t think we have fully explored strategies using it so you can’t expect others to know about it.

I am least grateful to know as apposed to just assuming we were being considered and also the reason we were not perceived as the best choice. Knowing this should help us at the championship.

Thanks Adam!

That’s another thing that really skewed the alliance results at MSC this year; the game. Roles were incredibly important as opposed to just how good a robot was; seeding didn’t matter as much as capabilities. This was also visible at Archimedes this year; the number 12 seeded team didn’t even move on to elims (sorry, 639 :/). Any teams that didn’t make it very far this year but still felt they had a good robot shouldn’t feel bad. A lot of good robots just simply didn’t bring anything new to an alliance so they weren’t picked.

I’m interested in further analyzing this data for this past season in addition to previous years, however I am missing data for post week-6 rankings for 2010-2014. Does anyone happen to have these spreadsheets available or know who I can contact to get them?

As far as I’m aware, that data is not available in the format you’re asking for it. However, you could tally the points for each team from the individual district sheets.

It’s generally easier to subtract the championship points then add up district points.

Sure. Either way it’d take a bit more work to reconstruct the tiebreakers, but certainly it’s all public data. For the future, it would be nice to have a separate sheet in the workbook with the pre-MSC rankings.

For Basel and anyone else who is interested, here are the spreadsheets for the 4 years I was doing the Michigan State Rankings. These files include ranking before state championship. I always keep a copy of the spreadsheet before State Championship just in case.

2014_Results.xlsx (184 KB)
2013_Results.xls (295 KB)
2012_Results.xls (313 KB)
2011_Results.xls (269 KB)


2014_Results.xlsx (184 KB)
2013_Results.xls (295 KB)
2012_Results.xls (313 KB)
2011_Results.xls (269 KB)