No impeachmen­t? Long shot

Some gam­blers buy stock on out­come of cur­rent events

Baltimore Sun Sunday - - BUSINESS - By Thomas Heath

Pres­i­dent Don­ald Trump of­ten re­gards the stock mar­ket as a re­flec­tion, equat­ing its ex­tended bull run as the ul­ti­mate af­fir­ma­tion. But the same free mar­ket levers that power Wall Street can also serve as a mov­ing ref­er­en­dum on his pres­i­dency and all things Wash­ing­ton.

In an era where you can bet on the name of the next royal baby or whether Elvis is still alive, so-called pre­dic­tion mar­kets like Pre­dic­tIt are po­si­tion­ing them­selves as se­ri­ous fore­cast­ers of ev­ery con­ceiv­able po­lit­i­cal out­come.

Pre­dic­tIt is a cur­rent events stock ex­change that poses hun­dreds of ques­tions — each its own “mar­ket” — at any given time. Users can plunk down as much as $850 to tap the next jus­tice to leave the Supreme Court (Ruth Bader Gins­burg at 73%), fore­cast whether Wil­liam Barr will still be at­tor­ney gen­eral at year’s end (92% say yes) or en­vi­sion a 2020 pres­i­den­tial run by Dwayne “The Rock” John­son (a hard no at 96%).

Then there’s Trump, whose impeachmen­t bat­tle has dom­i­nated the site. The odds of the Demo­crat-run House of Rep­re­sen­ta­tives vot­ing to im­peach him stood at 78% Wed­nes­day morn­ing, just ahead of the start of public hear­ings; they’d been as high as 80% in Oc­to­ber. The site doesn’t see the Repub­li­can-held Se­nate vot­ing for con­vic­tion.

John Aris­to­tle Phillips, the Wash­ing­ton­based com­pany’s co-founder, con­tends that the very na­ture of such plat­forms is a bet­ter pic­ture of what peo­ple are think­ing than what poll­sters can ex­tract from sta­tis­tics.

“When peo­ple have a lit­tle skin in the game, they are more likely to dis­till facts from fiction and dis­count fake news,” he said.”The the­ory here is that mar­kets can be ef­fec­tive at pric­ing fu­ture risk.”

An as­set’s value is based on the size of the trad­ing pool. Traders, much like their Wall Street brethren, make wa­gers based on their in­ter­pre­ta­tion of news re­ports, public state­ments and other fac­tors that can af­fect per­for­mance.

Pre­dic­tion mar­kets seek to har­ness the “wis­dom of the crowds” — the the­ory be­ing that large groups of peo­ple are col­lec­tively smarter than ex­perts and polls.

Though Pre­dic­tIt’s spokesman Will Jen­nings says it’s dif­fi­cult to make a “blan­ket ac­cu­racy assess­ment, given the va­ri­ety of mar­kets,” he said the Wash­ing­ton, D.C.based site’s hand­i­cap­pers are cor­rect 70% to 80% of the time.

It fore­cast, for ex­am­ple, that Brett Ka­vanaugh would be on the short­list to suc­ceed Jus­tice An­thony Kennedy on the Supreme Court. It also sig­naled that Ka­vanaugh would get the nom­i­na­tion hours be­fore it was for­mally an­nounced.

Pre­dic­tIt was on the wrong side of Brexit, when the United King­dom in­voked its sep­a­ra­tion from the Euro­pean Union in a let­ter from Prime Min­is­ter Theresa May. Nor did it call Trump’s 2016 elec­tion vic­tory; it said Demo­cratic ri­val Hil­lary Clin­ton had a 95% chance of win­ning Florida, for ex­am­ple, but she didn’t.

Pre­dic­tIt pegs Trump’s odds of re­elec­tion at 41%, which makes him the fa­vorite one year out. As the in­cum­bent, Trump has the his­tor­i­cal ad­van­tage.

Demo­cratic Sen. El­iz­a­beth War­ren, mean­while, has an 18% chance, for­mer Vice Pres­i­dent Joe Bi­den is at 13% and South Bend, In­di­ana, Mayor Pete But­tigieg stands at 12%, ac­cord­ing to the web­site.

Though Pre­dic­tIt comes across as quiz show meets Cae­sars Palace sports­book, the 5-year-old site doesn’t have book­ies and odd­s­mak­ers like the sports gam­bling par­lors that pop­u­late Las Ve­gas or lo­cal race­tracks. And it’s not a tra­di­tional poll­ster.

Rather, it’s a gam­bling-polling hy­brid that es­sen­tially asks peo­ple to back up their po­lit­i­cal prog­nos­ti­ca­tions with money.

Pre­dic­tIt al­lows traders to make pre­dic­tions on fu­ture po­lit­i­cal events by buy­ing 1to 99-cent shares in bi­nary out­comes. The re­sult­ing share price can be read as the prob­a­bil­ity of a spe­cific event oc­cur­ring, based on the col­lec­tive in­tel­li­gence of the mar­ket par­tic­i­pants, ac­cord­ing to the com­pany.

The odds are set by the traders and what they are will­ing to pay. If a con­tract that pays $1 if “Can­di­date X” wins an elec­tion is cur­rently priced at 53 cents, the mar­ket says “Can­di­date X” has a 53% chance of win­ning and 47% chance that they will lose.

Pre­dic­tIt’s team for­mu­lates ques­tions that must have two sides to the bet, known as a “con­tract.” Only one op­tion can be cor­rect, and it’s a dol­lar, win­ner-take-all.

Po­lit­i­cal con­sul­tant Pratik Chougule is the au­thor of the 2016 book “How to Make Money from Po­lit­i­cal Pre­dic­tions.” He agrees that pre­dic­tion mar­kets are more valu­able than pun­dits, talk­ing heads and politi­cians when it comes to dis­cern­ing po­lit­i­cal out­comes. But Chougule said Pre­dic­tIt has deep im­per­fec­tions due to a de­mo­graphic dom­i­nated by young, highly ed­u­cated men.

They are pre­dis­posed, for ex­am­ple, to grav­i­tate to a pres­i­den­tial can­di­date like An­drew Yang — a suc­cess­ful en­trepreneur who’s polling at 3% — as op­posed to life­time politi­cians like Bi­den or Sanders.

“If mar­kets are so ra­tional, why are they putting such a high like­li­hood of An­drew Yang or Hil­lary Clin­ton win­ning the Demo­cratic nom­i­na­tion?” Chougule said. “The mar­ket is over­stat­ing their odds be­cause the peo­ple who are bet­ting on po­lit­i­cal mar­kets are not any­where near a ran­dom sam­ple of the pop­u­la­tion.”


Pres­i­dent Don­ald Trump’s odds of re­elec­tion are pegged at 41% on Pre­dic­tIt, a cur­rent events stock ex­change.


Pre­dic­tIt has Ruth Bader Gins­burg at 73% as the next jus­tice to leave the U.S. Supreme Court.


Dwayne “The Rock” John­son as a 2020 pres­i­den­tial can­di­date? It’s a hard no at 96% on Pre­dic­tIt.

Newspapers in English

Newspapers from USA

© PressReader. All rights reserved.