MIT Invitational

[slowpoke]

Troy destroyed everyone in terms of the test, according to the supervisor.

Our B team got a mass score of 1860, time of 12.4 ==> 47.4435 on build. Our A team got a full mass score and a time score of 9 ==> 34 on build

Our A team got a 35 on test, our B team got a 26.

B team ended up 7th, A was 8th.

therefore 47.4435 + 50* 26/x is slightly larger than 34 + 50* 35/x (let's say the difference is 2 points, which is probably being somewhat generous)

Thus the max score is between 35 and 39

Could you explain how you got the 35-39 range for the test? Quickly looking on Wolfram Alpha, it seems like the max test score just has to be more than 33ish

If you simply solve the equation 47.4435 + 50* 26/x = 34 + 50* 35/x, you would get x = 33.47.., but by assuming the difference between the two teams is 2 points, you solve 47.4435 + 50* 26/x - 2 = 34 + 50* 35/x, which gives you x = 39.32... However, with one of these teams getting 35 points on the test, the minimum of the max score must be 35. At least that's what I think the thought process is.

Plus, since Troy destroyed everyone in terms of the test, I would guess the max score to be in the 40s at least (which you will get solutions to if you assume that the point difference between the two AB teams is greater than 2). For instance, if you assume a point difference of 3, the minimum for the max test score already becomes 44 (round up) and increases more as this point difference increases.

[Adi1008]

If you simply solve the equation 47.4435 + 50* 26/x = 34 + 50* 35/x, you would get x = 33.47.., but by assuming the difference between the two teams is 2 points, you solve 47.4435 + 50* 26/x - 2 = 34 + 50* 35/x, which gives you x = 39.32... However, with one of these teams getting 35 points on the test, the minimum of the max score must be 35. At least that's what I think the thought process is.

Plus, since Troy destroyed everyone in terms of the test, I would guess the max score to be in the 40s at least (which you will get solutions to if you assume that the point difference between the two AB teams is greater than 2). For instance, if you assume a point difference of 3, the minimum for the max test score already becomes 44 (round up) and increases more as this point difference increases.

Well, I already got 35 on the test... The 39 is because we're guessing that the scores were reasonably close.

Thanks for the explanation I was under the impression that the 39 was some sort of hard limit.

[kenniky]

If you simply solve the equation 47.4435 + 50* 26/x = 34 + 50* 35/x, you would get x = 33.47.., but by assuming the difference between the two teams is 2 points, you solve 47.4435 + 50* 26/x - 2 = 34 + 50* 35/x, which gives you x = 39.32... However, with one of these teams getting 35 points on the test, the minimum of the max score must be 35. At least that's what I think the thought process is.

Plus, since Troy destroyed everyone in terms of the test, I would guess the max score to be in the 40s at least (which you will get solutions to if you assume that the point difference between the two AB teams is greater than 2). For instance, if you assume a point difference of 3, the minimum for the max test score already becomes 44 (round up) and increases more as this point difference increases.

Well, I already got 35 on the test... The 39 is because we're guessing that the scores were reasonably close.

Thanks for the explanation I was under the impression that the 39 was some sort of hard limit.

Oh, no, I assumed the difference was 2 points because I figured that it'd be somewhat close, the more leeway you allow the larger the potential max score is (although I would put the difference at no more than 5)

[peter_ni]

Our B team got a mass score of 1860, time of 12.4 ==> 47.4435 on build. Our A team got a full mass score and a time score of 9 ==> 34 on build

Our A team got a 35 on test, our B team got a 26.

B team ended up 7th, A was 8th.

therefore 47.4435 + 50* 26/x is slightly larger than 34 + 50* 35/x (let's say the difference is 2 points, which is probably being somewhat generous)

Thus the max score is between 35 and 39

Could you explain how you got the 35-39 range for the test? Quickly looking on Wolfram Alpha, it seems like the max test score just has to be more than 33ish

Well, I already got 35 on the test... The 39 is because we're guessing that the scores were reasonably close.

What types of difficult questions were there on the test? So far, at regionals and the invitationals that we've went to, the tests have all been obscenely easy... I hope that the states' test will be a little more challenging to spread out scores.

[jkang]

Oh, no, I assumed the difference was 2 points because I figured that it'd be somewhat close, the more leeway you allow the larger the potential max score is (although I would put the difference at no more than 5)

Well, I already got 35 on the test... The 39 is because we're guessing that the scores were reasonably close.

Iirc I talked to the supervisor right before he went in for score counseling and he said Troy won by a 17-point margin (/100), or somewhere along the lines of that. So I'd imagine the scores were not actually that close.

[antoine_ego]

Oh, no, I assumed the difference was 2 points because I figured that it'd be somewhat close, the more leeway you allow the larger the potential max score is (although I would put the difference at no more than 5)

Well, I already got 35 on the test... The 39 is because we're guessing that the scores were reasonably close.

Iirc I talked to the supervisor right before he went in for score counseling and he said Troy won by a 17-point margin (/100), or somewhere along the lines of that. So I'd imagine the scores were not actually that close.

Wow, that would be impressive, and would still technically fit the calculations.

[Unome]

The ones from MIT (now that I've finally gotten around to posting this):

WIDI (16th) - pretty great. A mixture of several different types of materials; paper folding/cutting, drawing, pipe cleaners, pins, and that ridiculous piece of string (not to mention the four-cuts chair fold, which I just skipped entirely). Whereas last year's was more about being difficult to build in place (with it resting on the end of K'nex rods & connectors in a weird orientation) this year's was more about just being straight-up difficult to describe.

DyPlan (2nd) - Compared to what I was expecting, the multiple choice seemed a bit easy; I finished it in five minutes and missed 1 or 2 out of 25. However, the meat of the test was in the free-response, with 10 shorter problems and 5 longer ones, several of them very difficult or obscure (gravity anomalies!).

Disease (13th) - Same as in previous years, a great test. Pretty long, although somehow we got through the whole thing without splitting it. The use of actual articles was interesting, and the question difficulty varied well.

Astronomy (14th) - As usual a good covering of the test material and such, but (as with last year) easier and shorter than I'd consider appropriate for the tournament. If we had had any math skills during the tournament, we would have finished with at least ten minutes remaining for the 10 or so points of really difficult material (as it was, we totally blanked and got exactly 6 points on math).

Microbes (11th) - I can't really comment on this much because I don't really understand the event very well (we basically guessed our way through 60% of the test). More knowledge-based than I was expecting from looking at past nationals test.

Remote (13th) - Still pretty good by my standards because of all the low quality tests out there, but this one was a little disappointing. It was all free-response (<- not a criticism), and focused a lot on very specific things (e.g. a bunch of points for interpreting a Planck curve, listing out a bunch of measurements for a specific instrument), without much actual map-reading Edit: image interpreting (realized I said the wrong thing after reading windu's post). Also was a bit short, but that might just be because of the way it was written (with questions that were hard to guess on, so we ended up basically skipping).

[antoine_ego]

Fyi, on the MIT tournament website, there is a link to the tests and answer keys for this year. However, it's password protected, and I can't seem to find it.

[windu34]

Since Unome did it:
Electric Vehicle (3rd): Not very well run - supervisor didn't issue impound tickets and checking of device specs was lenient at best. Apparently the target was 20 cm to the left with respect to the starting point and can displacement reference point...

Hovercraft (24): Very well run and very good test. My formula for calculating current failed to provide useful data so I had to rely on wit for the time score portion.

Invasives (15): Same supervisor as last year so no surprise the test was mainly ecology and not at all like any of the other tests I will take the rest of the year (including nats). Good speed practice and well written, just bad focus.

Remote Sensing (16): Overall well written, but could have definitely used some more difficult application-based questions. Math was very straightforward. Test had way too many "straight from the cheat sheet" definition questions and statistic questions of instruments/platforms. Not nearly enough image interpretation.

Robot Arm (5th): Very well run. Amazing setup and great job on the supervisor's behalf.

Wind Power (6th): Supervisor(s) did not quite have it together and our teams turbine had another teams checklist on it (despite a clear "C12" marking). Test was very good and math was thorough. The last section was kinda dumb - they walked you through the calculation of the betz's limit - very confusing trying to figure out how to actually answer each question

[Adi1008]

Fyi, on the MIT tournament website, there is a link to the tests and answer keys for this year. However, it's password protected, and I can't seem to find it.

Your coach/sponsor should have received an email with the password