81

**Studying for the LSAT / EarlCat's LSAT Goal Calculator**

« **on:**March 24, 2008, 06:19:47 PM »

Pacing is one of the most difficult things to manage on the LSAT. If we go too fast, we get sloppy and miss questions. If we go too slow, we run out of time and miss questions. At some point between blazing through everything, and carefully considering every answer choice eight times is a sweet spot--the ideal balance between speed and accuracy. Easy concept.

The hard part is figuring out where that sweet spot is. You just got through the args section. You managed to answer 20 questions but you only got 13 of them right. Should you speed up? Should you slow down? Is this as good as it gets?

I had a student once who was having a very difficult time. She literally attempted 25 questions on a section and got 10 right. I handed her a calculator. "You've did 25 questions in 35 minutes. How much time per question did you take?" I asked.

"1 minute 24 seconds."

"Okay. How much time per questions would you have if you only tried 15 questions?"

"2 minutes 20 seconds."

"Now think about this," I said. "If you took almost an extra minute to consider

Now 15 questions isn't necessarily her magic number, but it illustrates an important point: Managing your speed has the potential to increase your score

For a while I have been brainstorming a way to figure out exactly where the LSAT sweet spot it. Obviously, someone breaking 165 or 170 ought to be completing the test. Someone who can literally never get a question right is better off guessing on every question. The people in the middle are a bit more difficult to manage.

A friend and I worked for a long time trying to find a way to point people toward the ideal balance of speed and accuracy for their current skill level. With a little bit of math and game theory, we came up with a formula we call OFI. OFI is based on several assumptions that I believe to be relatively safe. In general:

I've incorporated this into a simple calculator designed to point you in the right direction regarding how to pace yourself on future practice tests. It uses data based on individual sections of your most recent test to generate a speed and accuracy goal for that section on your next test. If your accuracy is too low to justify your speed, it slows you down. If your accuracy is too high, it speeds you up.

Not everyone's speed/accuracy curve is identical, but in theory, your required adjustment on each test should be smaller and smaller until you are consistently reaching your pacing goals. Once you meet a goal, the calculator will challenge you with a slightly higher goal next time.

Please try it out on your next several tests and let me know how things go.

EarlCat's LSAT Goal Calculator

The hard part is figuring out where that sweet spot is. You just got through the args section. You managed to answer 20 questions but you only got 13 of them right. Should you speed up? Should you slow down? Is this as good as it gets?

I had a student once who was having a very difficult time. She literally attempted 25 questions on a section and got 10 right. I handed her a calculator. "You've did 25 questions in 35 minutes. How much time per question did you take?" I asked.

"1 minute 24 seconds."

"Okay. How much time per questions would you have if you only tried 15 questions?"

"2 minutes 20 seconds."

"Now think about this," I said. "If you took almost an extra minute to consider

*every single question*you attempted, do you think you might get*just one more*right?" Of course she could. And that's an extra raw point on her score. But that's not all. Given the 4:1 odds against guessing correctly, she should get an extra 2 points on those remaining 10 questions just by choosing D for "Don't know." By simply slowing down, she could raise her score by as many as 3 points. Hell, even is she only got 9 correct, she's still likely to come out ahead.Now 15 questions isn't necessarily her magic number, but it illustrates an important point: Managing your speed has the potential to increase your score

*even if you don't improve your skills.*For a while I have been brainstorming a way to figure out exactly where the LSAT sweet spot it. Obviously, someone breaking 165 or 170 ought to be completing the test. Someone who can literally never get a question right is better off guessing on every question. The people in the middle are a bit more difficult to manage.

A friend and I worked for a long time trying to find a way to point people toward the ideal balance of speed and accuracy for their current skill level. With a little bit of math and game theory, we came up with a formula we call OFI. OFI is based on several assumptions that I believe to be relatively safe. In general:

- There is an inverse relationship between speed and accuracy. To a point, increasing speed decreases accuracy. You will do better on 20 questions in 35 minutes than you would on 1000 questions in the same amount of time.
- The increased likelihood of a correct answer is somewhat proportional to the amount of extra time spent on that question. An extra 30 seconds is more helpful than an extra 5 seconds.
- A student will guess on any unanswered question. (Which he/she always should.)

I've incorporated this into a simple calculator designed to point you in the right direction regarding how to pace yourself on future practice tests. It uses data based on individual sections of your most recent test to generate a speed and accuracy goal for that section on your next test. If your accuracy is too low to justify your speed, it slows you down. If your accuracy is too high, it speeds you up.

Not everyone's speed/accuracy curve is identical, but in theory, your required adjustment on each test should be smaller and smaller until you are consistently reaching your pacing goals. Once you meet a goal, the calculator will challenge you with a slightly higher goal next time.

Please try it out on your next several tests and let me know how things go.

EarlCat's LSAT Goal Calculator