760 (Q49 V45) - Great Result - Warning to MGMAT Users

I am very happy with my result over the weekend after several months of preparing to various degrees. I used the OG books plus Manhattan GMAT book set & online materials. I won't dive too much into my approach, as it may not work for many others (short version: read OG & a few practice questions surrounding 2 OG practice tests, final month I then tackled each section one at a time with practice essays, IR sessions, Manhattan guide books, OG Quant & OG Verbal, and Manhattan CAT usage, one final MGMAT full practice CAT, lots of timed quant sprints), but I do want to post for the benefit of others regarding the MGMAT CATs.

I know this has been covered in other places, but to reinforce the message: MGMAT QUANT CAT is WAY!!!! harder than the real thing. If you are preparing for the GMAT now and using MGMAT and are worried because you're struggling - DON'T WORRY! Their CAT questions are much harder and take much longer than the real questions. Here is my evidence:

Two official practice tests: Q49 and Q45, finished all questions first time, came up short by 2 second test. 200+ Quant questions from the OGs over various times sessions, both easy and hard, completed in avg < 2 min/question, ~95% correct. Real GMAT: Q49, all questions completed.

Compared to 116 MGMAT CAT Quant questions completed in 3 full 75 min sessions, 2 30mins sessions: Q31, Q38 and Q38 in full sessions, overall 56% correct and overall average of > 2min 30sec. Never completed more than 30 of 37 questions in any of the 3 full sessions.

These are statistically significant differences in results. I wrote to MGMAT and they replied that they do review their results based on student feedback for MGMAT vs GMAT scores and that I am an outlier...but I only buy that to a certain degree. It shouldn't be possible that someone like myself, who is clearly a far-above-average Quant performer on real GMAT questions (and on the real GMAT itself), could struggle so mightily on MGMAT if their questions are truly representative. They need to strive for a smaller standard deviation, and those who should see the greatest divergence in scores should be those in the middle of the bell curve (who are harder to predict and have most room to improve), rather than top performers being made to look below average.

So, to sum up - don't sweat it if you are having a similar experience on MGMAT CATs but feel like you're doing fine on the OG questions and practice tests. Use MGMAT to practice, review what you get wrong and remember those key lessons for when they show up on real questions, use the extra CATs to similate real testing conditions...but don't get caught up in the scores or the timing. Do time yourself on OG questions and make sure you're averaging 2min or less there.

Good luck all!