In the spirit of #SkinInTheGame, Taleb’s idea that pundits should at least stake their reputations on the strength of their knowledge, last year I made some predictions about 2017.
As usual, I am calibrating my predictions by comparing the percentage of predictions I got right at each probability level versus their probability (e.g., for predictions at the 70% confidence level, perfect calibration would represent getting 7/10 of them correct). Predictions with a probability rating of less than 50% are converted to their inverse.
Correct predictions are left as is, while wrong predictions are crossed out. Additional comments are in bolded italics.
Predictions – World
World Affairs & Conflicts
- No major conflict (>50 deaths) in East Asia/SE Asia that involves China and/or the US: 95%.
- US will not get involved in any new major war with death toll of > 50 US soldiers: 80%.
- No major conflict (>50 deaths), except Donbass, in the former Soviet space: 90%.
- The big one here is Armenia vs. Azerbaijan, but because reasons, the likelihood of a new flareup is now considerably lower than last year.
Oil prices are higher than $60: 50%. WTI Crude is at $58.47/barrel as of 12/23/2017. Perfectly calibrated prediction.
- China’s GDP grows by 6%+: 50%. Close: 6.9% in Q1, 6.9% in Q2, 6.8% in Q3.
- China will have more top 500 supercomputers than the US throughout 2017: 70%. And by quite the margin, now.
- There are fewer European migrant arrivals by sea than in 2016: 80%. 170,317 in 2017 vs. 362,753 in 2016, 1,015,078 in 2015.
- Venezuela does not undergo sovereign default: 80%.
- Israel will not get in a large-scale war (i.e. >50 Israeli deaths) with any Arab state: 90%.
- North Korea’s government will survive the year without large civil war/revolt (>100 deaths): 95%.
- No new global temperature record: 90%. Correct, this was the third hottest year.
- If only because it will be hard to beat 2016. That said, I expect shipping in the Arctic to continue booming.
- Radical life extension will not be developed: 99%.
- Superintelligence will not be developed: 99%.
- No further large scale civil wars/revolts/revolutions in Middle East/North African countries not already so afflicted: 70%.
- No large scale civil wars/revolts/revolutions (>100 deaths) in USA: 95%.
- No large scale civil wars/revolts/revolutions (>100 deaths) in China: 99%.
- No large scale civil wars/revolts/revolutions (>100 deaths) in Russia: 95%.
- No large scale civil wars/revolts/revolutions in any EU country (>100 deaths): 90%.
Syrian Civil War
- Bashar Assad will remain President of Syria: 95%.
- IS no longer in control of Raqqa: 50%.
- IS no longer in control of Palmyra: 80%.
- IS no longer in control of Mosul: 95%.
- IS still controls some territory in Iraq and/or Syria: 90%. That’s a few bits of bare desert, but still.
- Syria still controls Aleppo: 80%.
- The first danger is a Turkish stab in the back, in which its proxies turn hostile and, in effect, encircle Aleppo between al-Bab and Idlib (though I don’t view that as being likely). The second danger remains the banal fact that most of the SAA is no good, especially its garrison units. I don’t fully exclude the possibility of the rebels seizing the city back once the elite units are sent off somewhere else.
- Syria still controls Deir ez-Zor airport: 90%.
- Idlib is still under rebel control: 70%.
- That province has the lowest polled support for Assad of any in Syria, and that number has correlated very well with the difficulties the Syrian state has had in reimposing its authority there.
- US/Allies will NOT impose no fly zone over Syria: 95%.
- Turkey will not “backstab” Russia and the Syrian government: 90%.
- The Syrian Civil War is still ongoing: 95%.
- Russian intervention in the Syrian Civil War is scaled down relative to today in a year’s time: 70%.
War in Donbass
- War in Donbass doesn’t reignite: 80%.
- Incidentally, and quite ironically, Trump’s election may well have made Ukraine’s position safer (that at least is also the opinion of Igor Strelkov). Were HRC to bloody Putin’s nose in Syria with a no fly zone, three guesses as to who the object of a “short victorious war” to restore Putin’s reputation might have been. With Trump, Syria as a source of escalation is removed.
- The Ukrainians also now have far fewer reasons to heat things up, because its a safe bet that Trump won’t be interested in pulling their chestnuts out of the fire.
- No “Putinsliv”/abandonment of Russian support for DNR/LNR, with Ukraine recapturing Donetsk and Lugansk: 99%.
- Mariupol still under Ukraine control: 90%.
- Dnipropetrovsk, Odessa, and Kharkov all still under central Ukrainian control: 95%.
- Kolomoysky has been defanged as a regional attractor, but Odessa remains a potential powderkeg. I think Ukraine will pull through – as I’ve always noted, Right Sector thugs count for more on the streets than protesting pensioners.
- Poroshenko remains in power: 90%.
- The Ukraine does not undergo sovereign default: 90%.
- The next key date is January 2017 when the Commercial Court in the UK rules on the status of Russia’s $3 billion loan to Ukraine.
- The Ukrainian economy shows GDP growth: 80%. Growth: Fell from 4.8% in Q4 2016, but remains decidedly positive at 2.5% in Q1, 2.3% in Q2, 2.1% in Q3.
- If only because there is very little room for it to fall any further, with GDP per capita only 70% that of the UkSSR in 1989.
- Putin remains Russian President: 95%.
- Putin announces entry into Presidential race for his second (fourth in total) term: 90%.
GDP grows by 2%+: 50%. Russia’s growth was 0.5% in Q1, 2.5% in Q2, and 1.8% in Q3; growth for the entire year will probably be around 1.7%; will need to be 3.2% in Q4 to eke out 2%.
- Putin’s approval rating according to Levada doesn’t dip below 70% during the year: 50%. They didn’t even dip below 80%; minimum was 81% in June and November.
Russians have a more positive view of the US than of the EU as of the last Levada poll in that year: 60%. This prediction actually came true by March 2017, when for a single month, marginally fewer Russians approved of the EU (36%) than of the US (37%) – a stunning reversal, since before the Ukrainian crisis, the EU had traditionally been seen in a much better light. Then came the attack on Syria, and by May, US approval was back down to 23% versus 27% for the EU. As of December, the EU has an approval rate of 28% to America’s 24%.
- There will not be any substantial anti-government protests (>10,000): 90%. The most substantial oppositionist protest was Navalny’s “He is Not Dimon” meeting at Tverskaya; I was there personally and estimated 7,000-8,000. Khodorkovsky’s “Enough” got a pathetic 200. This year’s Russian March likewise had almost no people, the event having long been hijacked by pro-Ukrainians. The one demonstration that did satisfy this condition was the Anti-Khrushchevki Demolition Protest, which gathered around 20,000 people – however, they did not present a united anti-government agenda, even to the point of sending Navalny packing, so I don’t think this counts.
- Trump isn’t interested, and the Europeans don’t have as much money.
Natural population growth: 70%. This year saw an ~11% decrease in Russian birth rates, a strange and sudden collapse that seems to have been part of a general Eurasian (Ukraine, Baltics) and maybe even industrialized world trend. Preliminary figures for Jan-Oct 2017: 1,418,090 births (1,588,095 in 2016), 1,532,955 deaths (1,567,803 in 2016), translating to natural decrease of -114,865 (+20,292 in 2016). Looks like the definitive end of the brief period of Russia’s natural growth during 2013-15.
- Total population growth: 95%. Net immigration usually runs at around 250-300k / year, so this is still true.
- The Crimea remains Russian: 99%.
- Trump remains US President: 95%.
- Rex Tillerson becomes and remains Secretary of State: 80%.
- Had already predicted his rise at PredictIt well before Trump made it public (see right).
- Hillary Clinton does not get prosecuted: 90%.
US economy grows by 3%+: 50%. Growth: 2.0% in Q1, 2.2% in Q2, 2.3% in Q3 – not gonna happen. US relaxes or removes Russia sanctions: 50%. We’ve now transitioned to the US likely selling Javelins to the Ukraine, LOL.
- Trump might support that, but Congress surely won’t.
- The Alt Right acrimoniously splits into Trumpists and anti-Trumpists: 70%. Absolutely.
- This prediction actually dates back to May 2016.
- Incidentally, this is yet another fascinating Putin/Trump parallel – Putin’s Solovyev/Starikov are Trump’s Milo/Cernovich, while the ethnats have at best a “mixed” relationship with them.
- The “Ferguson Effect” reverses or at least stabilizes (homicides in major urban areas peak off): 60%. Apparently so.
- *** Freedom House lowers United States Freedom Rating: 50%. We need to wait until Jan 31 for the Freedom in the World 2018 report to come out; moving this prediction into next year’s.
- For instance, lowers “Civil Rights” category from 1 to 2, because they are too triggered by Trump.
- There will be fewer campus disinvitations “from the Left”: 70%. In 2017, there were 29 disinvitations (43 in 2016), of which 24 “from the Left” (35 in 2016).
François Fillon becomes French President: 70%.
- Marine Le Pen will not be French President: 80%.
- Yes, to be sure, last year’s events made fools of mainstream analysts, but the fact remains that the French political system is very tough for nationalists. Trump wouldn’t have won in that format either.
- Merkel remains German Chancellor: 70%.
- Frauke Petry will not be German Chancellor: 90%.
- Based on opinion polls, I just don’t see how it’s possible.
- No country leaves the Eurozone: 95%.
- Article 50 is invoked in the UK: 90%.
- No second Scottish referendum is called: 80%.
- Scotland remains in the UK: 95%.
- No Islamic terrorist attack in Europe causing more than 100 deaths: 70%.
The EU relaxes or removes Russia sanctions: 60%.
- The Mediterraneans don’t care and are getting increasingly restive.
The Unz Review has fewer viewers than in 2016: 70%. I was wrong. Great!
- Would be great to be wrong, but it’ll be hard to beat the (last) Current Year.
Mount & Blade: Bannerlord released: 95%. GRRM publishes Winds of Winter: 80%.
Predictions – The AK
- I will still be in Russia: 95%.
I will make my hajj to Crimea: 50%.
- I will continue blogging: 95%.
- I will write a record amount of blog posts (172+): 70%. Will finish up at around 260, versus 128 in 2016 and 130 in 2015.
- I should imminently have a lot more free time so my productivity should increase.
- There will be more comments than in 2016: 80%. 17,743 and counting, versus 6,204 in 2016 and 5,504 in 2015.
- There will be more visits and views than in 2016: 80%. 751,826 views and counting, versus 493,156 in 2016 and 382,574 in 2015. I’ll be sure to do some more statistical navel-gazing on the 10th anniversary of my blogging career (January 9, 2018). How do I stack up against the others here? Steve Sailer gets an order of magnitude more views (c.11 million in 2017). James Thompson will be a little shy of half a million.
- I will create a Russian language blog and write more than 10 posts for it: 60%. I actually wrote considerably more for akarlin.ru – around 32 posts. Though due to low intensivity, commentary is inactive. That said, our ROGPR podcast is growing by leaps and bounds – the YouTube channel already has 4,000 subscribers.
I will write 30+ book reviews: 50%. Total fail.
- Reviewing things will definitely be more of a priority from now on.
I will write 5+ game reviews: 50%. Total fail.
- I will write fewer than 5 movie reviews: 80%.
I will author or coauthor a paper: 90%. Total fail. Though a couple are now tantalizingly close.
- An S Factor analysis of Russia is forthcoming.
I will author or coauthor two papers: 70%.
- I will not author or coauthor more than two papers: 80%.
I will finish writing at least one book: 70%. Total fail. I will publish that book by Dec 31, 2017: 50%. Total fail.
- I will not be banned or shadowbanned on Twitter or Facebook: 90%.
- Of course it helps that I barely use them nowadays.
I will finally get a RationalWiki “hagiography” ala JayMan or hbdchick: 50%.
Here is my calibration graph:
Did pretty poorly on the 50% confidence predictions. However, of my 50% confidence personal predictions, all five were wrong; cancel them out, and the calibration for this level would be a much more respectable 42%.
My main weak point was being massively over-optimistic on what I could get done in 2017.