PredictWise Blog

Microsoft Research’s Indian Election Game

Bookmark and Share

The final voting for the India’s general election concludes at 6:30 PM local time on Monday, May 12 and, “as the voting closes in India, no one knows what the results will be.” There are total of 814 million eligible voters in this massive display of democracy, which is why voting is spread out over more than a month (not enough polling machine and election workers, so they need to move around the country). It will take an additional four days to count the final vote tallies, due to be released on Friday, May 16, although we expect exit polls to be released as the voting closes.

While readers from the United States are accustomed to intense and accurate coverage of the election polls through Election Day (the US is already inundated with 2016 election polls), India bans polling from being released during the voting portion of the campaign. Microsoft Research saw this as an opportunity to test a new crowd sourcing prediction game to see how well we could do at predicting the unpredictable.

The Indian Elections Game, joint work between Microsoft Research India, including Shipra Agrawal and Nikhil R. Devanur, and Microsoft Research New York City, including Miro Dudik and myself, was released just before voting began on April 7 and ran through today’s final voting day. Nearly 4,000 users created their predictions on the number of seats that any given party, or combination of parties, would get in any of the 35 voting states and union territories. For example, a user may look at Delhi’s seven seats and predict that BJP would get between 4 and 5 seats while Congress (INC) would get between 0 and 1 seats.

As the crowd makes predictions on the outcome of all 35 voting states, the national parties, and national alliances, the backend of the game examines the crowd’s predictions and creates the probability of any given outcome occurring. In this example, since game’s aggregation of the crowd’s predictions currently concludes there was just less than 50% likelihood of this combination occurring, a prediction that wagered 1 point on BJP getting 4 or 5 seats and Congress getting 0 or 1 seat would yield 2.28 points if correct and 0 points if incorrect.

Computing the probability was not trivial, as the game had to learn how the different parties interacted within a given district and between districts. And, unlike the United States where there are just two major parties, there is large number of viable parties, some national and some regional, competing for the 543 seats. The potential number of outcomes is huge! In future blog posts, over the next few days, we will go into more depth about how all of that was constructed.

But, right now, just as the election closes, and the exit polls are released, we wanted to share with you the final probable outcomes that our game generated from the approximately 4,000 users who played over the last month.

We expect the opposition party BJP to carry 228 of the 543 seats and their national alliance, the NDA, to carry 270 seats. The party currently in power, Congress is expected to carry just 87 seats and its alliance, the UPA just 110. The Third Front alliance is expected to carry 80 seats. That leaves 83 seats to smaller parties that are not formally aligned with any of the three major alliances. The national party predictions have a margin of error plus or minus 7% of seats and the national alliances plus or minus 10% of seats.

This prediction is generated by considering each voting states independently and then considering how they aggregate up together. Below we list each of the 35 states and our individual predictions:

Please check back over the next few days we discuss this game and its predictions in more detail!

Eurovision Finals - Live

Bookmark and Share

7:00 PM: Sorry to get the UK's hope up; only 2 of my 10 did not finish in the top 10! But, stoked that Austria, Sweden and the Netherlands performed to expectations. And, nice to Armenia hang in there, even if they never really regained their mojo after their semi-final round. Russia's reasonably high finish was more political than anything else ... it was getting really testy in that arena whenever anything from Russia was mentioned. Until next year ...

6:20 PM: Happy to see top three going into finals finish as top three, but Austria did leapfrog Sweden!

6:17 PM: It is going to be Austria, the Netherlands, and Sweden ...

6:14 PM: Only Netherlands and Sweden are still alive after 25 of 37 votes ... it is Austria's to lose still ...

6:03 PM: Wow!!! Sweden gave their top points to Austria ... good sports!

5:58 PM: Voting is 1/3 through and Austria and Sweden are in a tight fight for the top spot ... 75% says it is one of them! 44% for Austria and 31% for Sweden ...

5:48 PM: Voting is finally being released and crowd had to be reminded not to boo Russia ...

5:22 PM: Still voting, but the market is getting more and confident ... Betfair almost to 50% that Austria will win this one ...

5:09 PM: Here are the updated likelihoods, post-finals ... Austria goes into the voting at 39% to Sweden's 24%, and the Netherlands 13%.

4:57 PM: Netherlands was good, but not great. Going into the final competitor we have Austria at 36%, Sweden at 26%, and Netherlands at 13%. The UK is up now at 6% ...

4:45 PM: Final contenders are about to take the stage and Austria is still in the lead with 32% and Sweden with 25%. The Netherlands at 16% is up next though and the UK at 5% is last ...

4:18 PM: Everything is holding petty steady as we wait for the last of the big five in about 20 minutes ...

4:11 PM: Sweden did well, but not well enough to get that lead back from Austria. Current forecast is Austria at 34% and Sweden at 26%. The Netherlands and UK are the 24th and 26th competitors respectively (out of 26). So still waiting on them ...

4:05 PM: It is 4:05 PM in New York City, USA or 10:05 PM in Copenhagen, Denmark and I am now blogging live for the rest of the Eurovision competition. Halfway through tonight's finals there is some serious movement in the likelihoods of victory. With Armenia and Austria both already finished the crowd has spoken - Austria rocked and Armenia flopped. Austria is now at 34% likelihood for victory with Armenia down to 3%. Sweden is singing now ...


Eurovision Finals

Bookmark and Share

There has been an impressive amount of movement over the last week as the world (or is it Europe?) watched 20 of the 26 finalists compete their way through the semi-finals.

1) Three competitors shined in the semi-finals: Sweden moved from 21% to 25% likely to win the contest. Sanna Nielsen's Undo has 1.2 million YouTube views, nearly 400,000 coming since the first semi-finals. Austria and Netherlands are the more exciting story. Austria has shot from 2% to 21% after Conchita Wurst's Rise Like A Phoenix performance in the second semi-finals. It is worth a view for those of you who are not going to watch tonight ... it already has 4.8 million YouTube views. Similarly, the Netherlands moved from 1% to 18% after The Common Linnets' Calm After The Storm first semi-final performance.

2) Armenia holds on as the fourth most likely winner at 10%. After coming into the contest as the heavy favorite, Aram MP3 - Not Alone did not stun in the first semi-final, but 3.8 million YouTube views, with 1 million in the last few days since the semi-finals, shows they are still generating a lot of interest.

3) Rounding out the top five is Great Britain. As an automatic qualifier, the crowd did not get to see Molly Smitten-Downes' Children of the Universe. Thus, it is not surprising that it held on at 8% likelihood of victory.

4) Five competitors cluster between 6 and 10 on my list: Ukraine (4%), Hungary (4%), Denmark (3%), Greece (1%), and Spain (1%).

5) The remaining 16 are extremely unlikely to win.

Eurovision Semi-Final 2

Bookmark and Share

5:30 PM ET: We had another strong night tonight with our semi-final 2 predictions. All 5 of our "Very Likely" finalists breezed through the semi-finals. 2 of 3 of our "Likely" finalists won, with Israel not making it to the final. 1 of 2 "Bubbles (in)" made it with Poland in and Ireland out. 2 of 3 "Bubbles (out)" got to the finals with Lithuania missing the cut and Belarus and Slovenia triumphing. Finally, both of our "Not Likely" did not make it ...


1) Likely winner: the likely winner looks dramatically different between the two semi-finals for two reasons. First, 10 competitors who previously had to pass through the semi-finals are now in the semi-finals … getting over that barrier increases their chances of victory. Second, fans got a close look at those 10 competitors on Tuesday and it changed some rankings.

Sweden has overtaken Armenia as the most likely winner at 25% to Armenia’s 18%. Clearly Armenia’s semi-final performance was not as strong as expected as it went into the semi-finals with 32% likelihood of victory. The surprise of the day was the Netherlands that pushed up to 14% with a strong semi-finals performance.

Expect one or two of the winners of tonight’s semi-finals to vault into competition, after they escape into the final round! More details below tonight's preview ...

2) Semi-final 2: Tonight's semi-final will send 10 of 15 competitors to Saturday's finals. Here are the live predictions. This semi-final is a lot tighter than Tuesday’s semi-final with no lock and no definite misses.

Very Likely: Norway, Greece, Romania, Austria, and Malta are all over 80%. Unlike Tuesday’s semi-finals there are no locks this time!

Likely: Finland, Israel, and Switzerland are all about 75% to make it to the finals.

Bubbles (in): Poland and Ireland round out my top 10 with 67% and 62% likelihood respectively.

Bubbles (out): Belarus, Slovenia, and Lithuania are the likely competitors to pick up one or two spots in the finals at about 50% likely each.

Not Likely: Macedonia and Georgia are 22% and 14% each. Again, this is not likely, but I am not saying they are impossible.

3) Likely winner of Eurovision 2014 from before the first semi-final and now before the second semi-final:

Eurovision Semi-Final 1

Bookmark and Share

5:21 ET: All good for our first set of predictions! All 5 “locks” made it to the finals, along with 2 of 3 “likely” and 1 of 2 “bubbles (in)”. The two surprises came from 1 of 2 “not likely” and 1 of 3 “very longshots” … the 1 bubbles (out) did not make it.


I will have much more to say about this later in the week, but here is my first salvo on Euorovision 2014.

1) Likely winner: I have two clear favorites going into the contest. First, is Armenia at 31.7%. Their video, Not Alone, has been viewed 2,808,965 times on YouTube. Second, is Sweden at 21.1%. Their video, Undo, has been viewed just 820,508 time on YouTube. Finally, the most likely winner from the big European countries, Great Britain is third at 9.9%. Children of the Universe has been viewed 545,016 times on YouTube.

2) Semi-final 1: Tonight's semi-final will send 10 of 16 songs to Saturday's finals. Here are the live predictions.

Locks: Armenia, Sweden, Azerbaijan, Ukraine, and Hungary are virtual locks to make it to the finals.

Likely: Russia, Netherlands, and Belgium are all very likely at over 75% to make it to the finals.

Bubbles (in): Montenegro is 66% and Estonia is 64% to make it to the finals, rounding out the top 10 most likely.

Bubbles (out): Moldova is 45% to squeeze into the finals.

Not Likely: Iceland and Albania are possible, but long shots at about 25% to make it into the finals.

Very Longshots: Latvia, San Marino, and Portugal are all very longshots at less than 20% to make it into the finals.