GMAT Tip of the Week: GMAT Scoring – Best of 7

GMAT Tip of the WeekWith the Major League Baseball playoffs on many minds, and the beginning of the NBA and NHL seasons on others, you’ll hear a lot in the news these days about trends in a “best of 7” series, in which a team needs to win four games to advance to the next round.

“Only x% (a very small percentage) of teams have ever come back from a 3-1 deficit to win”
“Only one team has ever come back from a 3-0 deficit to win”
“If a team takes a 2-0 lead there’s a (very high) chance that they win the series”

These adages go much like the trends you see thrown around in the GMAT space:

“The first 10 questions have a disproportional bearing on your score” (note, this is an overplayed rumor)
“If you get the first 10 right you’re guaranteed a…”
“If you get the first 10 wrong you’re guaranteed…”
“We’ve researched it, and if you get the first 10 right vs. the middle 10 right vs. the last 10 right…”

And in either case, people buy the hype without questioning it – or “critical reasoning” it. Think about a 162-game baseball season for a team like the LA Dodgers, who went down in the NLCS by a 3-1 margin. The Dodgers from that point had to win 3 games to advance (they’ve won the first of those and play again tonight), a herculean feat if you believe the SportsCenter anchors and newspaper pundits, but wait – the Dodgers won 3 straight games dozens of times this season. They won well more than 3 in a row several times. Is winning 3 in a row really as big a deal as they say?

The major reasons that it’s so hard to come back from 3-0 or 3-1 are about the same as the reasons that the first 10 questions of the GMAT are so predictive of your score:

1) Losing the first 3 – or answering several of the first 10 questions incorrectly – are a sign of being overmatched.

People forget in sports that it’s not at all uncommon to win 3 or 4 straight. It’s not the daunting nature of *that* feat that makes coming back so hard – it’s doing it against a team that has demonstrated that it’s better than you. You have to beat a team that’s been repeatedly beating you – they’re probably better. Similarly on the GMAT, it’s not that the first 10 questions “matter more” toward your score, it’s more that if you get 8 of them right, you’re probably really good at this GMAT thing, and if you get 7 of them wrong you’re probably not, at least not today. So yes, if you analyze practice tests those who do better in the first 10 almost always score better than those who do worse, but it’s not the order of the questions that does it, it’s what the results are starting to prove – if you don’t have it, you don’t have it.

2) Losing the first 3 – or answering several of the first 10 incorrectly – lower your margin for error.

Another major reason that going down so early in a series, or starting so slowly on the GMAT, tends to correlate with poor results is that in order to recover you have to be a lot closer the rest of the way – you can’t afford mistakes. If you’re down 3-1 in a “4 wins and it’s over” series, you can only afford one more loss until it’s over. You just can’t make mistakes at that point – one ill-timed throwing error, one meatball pitch taken for a home run and even if you’re building momentum you’ve lost it.

It’s similar on the GMAT – you *can* recover from a poor performance on the first 10 questions, but very few do. It’s rare for any student to get 10+ questions right in a row at any point, partially because the adaptive system continues to challenge your upper threshold the more you get right, but also because you’re human…you make mistakes. If your estimated score is lower than you’d like after the 10th question, you may just fall victim to the old sports adage “they didn’t lose, they just ran out of time” – a phrase that applies to teams that try to mount a ferocious comeback and fall just short. If on the quant section you’re behind your goal after 10 questions, you only have 27 more to build that up, and like an itsy-bitsy spider climbing up a water spout, if you slip down a notch or two because of an ill-timed mistake (a slippery spout?) you have that much farther to climb.

Now, all of this serves to show that the myths about GMAT scoring tend to – in classic GMAT Critical Reasoning style – mistake correlation and causation. But there is another actionable lesson here:

You should:

Avoid mistakes in the first 10 questions, even if that means spending a couple extra seconds to double-check your work. Like in best-of-sevens, early struggles dig you a very deep hole that’s tough to climb out of, so it’s important to avoid early “losses” if at all possible. The first 10 questions are worth an extra 90-120 seconds of your time (in total) to make sure you don’t set yourself up for failure.

You should not:

Go “all-in” on the first 10. This is where students succumb to the myth. Say that the Detroit Tigers, having won game 1 of their best-of-seven with Boston, looked at the stats and said “if we win the first 2 on the road we’re all but guaranteed to win the series”. They might, then, have replaced starter Max Scherzer (who gave them 7 great innings before turning it over to the bullpen) with aces Justin Verlander and Doug Fister – the Tigers could have won Game 2 by burning out their superstar arms in relief…but then they’d have had no one left for Games 3, 4, and even 5.

This is what many GMAT students do – they spend 50% more time than they should, or more, on the first 10 questions, because “if you get the first 10 right you almost always end up with a 700+”. That’s mainly true because of point 1 above – those who get 8 to 10 of those first 10 right are usually great test-takers. If you game the system and set yourself up for failure in the last 30 questions, you’re using false correlation and mythology, and you’ll almost always get burned.

So as you watch these best-of-sevens unfold and hear the pundits talk about statistical trends, heed a lesson about GMAT scoring – what you do early does matter a lot, but much more as an indicator of how you’ll perform throughout and less as a direct causation of success. Or as Yogi Berra said best – and we’re not sure if he meant “baseball” or “GMAT” – it ain’t over ’til it’s over.

Plan on taking the GMAT soon? Try our own new, 100% computer-adaptive GMAT practice test and see how you do. And, be sure to find us on Facebook and Google+, and follow us on Twitter!

By Brian Galvin