What did the pros get right about Goblins vs. Gnomes?

Sometime in the next week (or so), Hearthstone‘s latest expansion, The Grand Tournament, will be unleashed. With 132 new cards, pros and pundits (including here at the Daily Dot) are falling over themselves to give their opinions on cards based purely on sight and theorycrafting.

Something very similar happened around nine months ago, ahead of the release of the 123-card Goblins vs Gnomes expansion. It was the first large card set added into to the game since release, and it has changed the game in ways that are felt to this day. Can you imagine playing Hearthstone without seeing cards like Piloted Shredder, Antique Healbot, and Dr Boom every day?

Building upon the work of Reddit user “p_red”, who analyzed the predictions of three of Hearthstone’s top personalities (Octavian “Kripparrian” Morosan, Brian Kibler and Jeffrey “Trump” Shih), we’ve included the reviews of six other pros: Kacem “Noxious” Khilaji, Nathan “ThatsAdmirable” Zamora, Jason “Amaz” Chan, Jan “Ekop” Palys, Keaton “Chakki” Gill and Brent “Backspace” Kaskel.

Using the methodology suggested by another reddit user, “slate15”, we grouped the predictions into three categories—correct, false positive (that they predicted the card would be rated as “good” when it didn’t turn out that way), and false negative (the opposite). The cards were rated simply as good or bad, based on “p_red”‘s original determination of whether or not the cards appear in decks at the top of the meta.

Overall, the pros actually faired very well. All nine predicted more than 55 percent correctly, with six getting at least around two out of three right. You can see the full data set here.

How did the pros fare overall?

In the original Reddit investigation, Brian Kibler fared the best, with a 70 percent accuracy rate. Kibler correctly predicted 84 out of the 120 cards he reviewed, with 25 false positives (the second lowest) and 11 false negatives. Some examples of cards he correctly predicted out of step with the rest were Crush and Ogre Brute.

Kibler was topped by one other pro, however. Keaton “Chakki” Gill scored an impressive 74.78 percent, getting 86 out of 115 predictions right.

At the opposite end, Kacem “Noxious” Khilaji was bringing up the rear with a 55.88 percent success rate—though in truth, his predictions were the most difficult to parse. An eternal optimist, Khilaji was able to see potential positive applications in most cards and present them as having potential. As such, he was the only pro to record zero false negatives, but posted 45 false positives. But he was also one of only two pros to predict the impact of Imp-losion on the game.

Jason “Amaz” Chan and Nathan “ThatsAdmirable” Zamora also posted scores under 60 percent.

What cards did the pros think would be good?

There were an average of 32 false positives per player, with only 11/35 good cards universally agreed upon.

Only one card was incorrectly predicted as good by all nine: Vol’jin. Many thought his utility as a removal and a strong mid-game minion would prove to be a strong presence, but Priest continues to be infrequently picked and, when it is, Vol’jin doesn’t feature. Another Priest card that was overrated by eight out of nine pros was Upgraded Repair Bot, as sadly mech Priest never took off.

Jeeves and Anodised Robo-Cub were also rated as good by all but one player.

In terms of what was predicted correctly, all players seemed to see the strength of many of the mechs which immediately overtook the meta following the GvG release. Mechwarper, Snowchugger, Goblin Blastmage, and Tinkertown Technician were predicted correctly by everyone, and went on to form the backbone of Mech Mage in early 2015.

What cards did the pros think would be bad?

Far fewer cards were incorrectly rated as bad by the pros we looked at, with five of them recording single figure numbers of false negatives.

The two biggest whiffs in this category are Dr Boom and Imp-losion. While people often suggest that almost all pros were wrong on Dr Boom, Morosan, Chan and Khilaji all correctly predicted it would see play in the meta.

Another card that most pros did not see as strong was Tinker’s Sharpsword Oil, the key card in the main Rogue deck in today’s meta. Instead of being a Deadly Poison with one extra attack and a hard to pull off combo, Oil has ended up as one of the strongest cards in the class.

Five of the pros also predicted that Annoy-o-Tron and Fel Reaver would not see play.

Image via Blizzard

If you liked the article, you’ll love our video. Check out today’s Quick Cast news update.