Real or whack? http://www.lsat-test-prep-nerd.com/ Forum

[chewdak]

Claims to have a statistics based method of guessing on the questions you don't know with 2/3 rather than 1/5 accuracy.


Edit: I re-read their claim: it is 63-70% better at guessing, not accuracy, which they then compare with accuracy for random guessing.

[UFMatt]

I haven't looked at the site, but they probably expect you to calculate the percentage of each answer choice for what you do know, then figure out which option is lagging behind. Maybe combine that with the question number and question position in relation to a subsection (e.g. reading comp or logic games), as those seem to have trends as well.

It'd probably take more time than it's worth to calculate.

[chewdak]

I haven't looked at the site, but they probably expect you to calculate the percentage of each answer choice for what you do know, then figure out which option is lagging behind. Maybe combine that with the question number and question position in relation to a subsection (e.g. reading comp or logic games), as those seem to have trends as well.

It'd probably take more time than it's worth to calculate.

Probably a waste of time one way or another.

[cubswin]

Looks like a load of crap. Maybe if you have a large number of questions on which you have absolutely no fucking idea and can't even eliminate an answer choice, this might work. Not that I know exactly what they're suggesting, but they say they have "discovered numerous patterns and correlations between question types and alphabetical answer choices."

Perhaps it'd be useful to people who guess blindly on half the questions.

[chewdak]

Their claim is "use our statistical answer choices and boost your chances of getting that question right by 63% to 70% compared to random guessing or choosing B which has a chance of only being 20% correct".
I misread this as claiming 2/3 accuracy on guessing, easy mistake to make as these numbers are right next to each other.
Actually, their claimed accuracy must be 2/3 over 20%, or about 1/3.
Pretty sneaky.

[abbas123]

whack

[amputatedbrain]

If you are having to guess so much that this interests you, you're not ready to take the LSAT.

You really shouldn't have to make a total guess more than once or twice. Most of the time you'll be able to eliminate at least some blatantly wrong answers, to at least 3, or usually 2, in which case your odds are then 50/50 anyway.

[Woozy]

You really shouldn't have to make a total guess [strike]more than once or twice[/strike] ever. Most of the time you'll be able to eliminate at least some blatantly wrong answers, to at least 3, or usually 2, in which case your odds are then 50/50 anyway.

FTFY

Their claims of 4-9 extra points for everyone (even from 170-179) by adopting a better guessing strategy are outrageous. Their claims that there may be some statistically significant patterns in the answer choices cannot be dismissed as easily, but I am inclined to remain skeptical unless convinced otherwise.