Resolves YES if at any point in time, Russia takes control of the train station "Bakhmut 2" in Bakhmut, Ukraine, by April 20, 2023 (Ukraine time), according to maps provided by the Institute for the Study of War (ISW).
If this area is shaded red on the ISW maps, the market will resolve YES. If this area remains shaded orange or not shaded for the entire time period, resolves NO. Resolution will be based on the ISW maps dated through April 20.
See https://manifold.markets/post/isw-ukraine-war-maps for context on how ISW maps are made, what they mean, and why ISW is a commonly considered a reliable map source.
The specific location used for this question is: on the ISW interactive map, the dot if you search "Bakhmut 2". The shading of the map over the center of the dot when fully zoomed in will be used for resolution.
(Other reliable media reporting may be used instead if for some reason ISW is unavailable.)
Resolution criteria based on https://insightprediction.com/m/169724/will-russia-take-bakhmut-by-april-20th-2023
(Just making the title more descriptive, no change to the question itself): Will Russia control Bakhmut by April 20? → Will Russia control central Bakhmut by April 20?
@nickten ISW has the advantage of having an archive of the past published maps. Does deepstatemap have that?
Can always use the orange shading instead of just the red shading if we want to include areas recently claimed.
I'm planning to create some more markets similar to this. What do people think are the pros and cons of the different maps sources that people look at?
Note that for ISW we can use the orange shading, which AFAIK updates about at the same pace as the other maps (unlike the red shading which is more conservative).
@jack Personally, I didn't particularly mind the procedure that was used this time. But maybe I'm biased 🙃
The original title was "Will Russia control Bakhmut by April 20?". Whether or not they control the train station now, they don't control Bakhmut as a whole (nor do the Ukrainians). The title is not what defines the market, not nonetheless this makes me feel as if the outcome is less "unfair".
I guess one solution could be to resolve "yes" if ISW shades something red and N/A if ISW shades something orange.
@TorBarstad I think it's ok, but it's preferable if the resolution criteria maps as closely as possible an intuitive question. For example, in this case the more intuitive statement of the question might be "Will Russia have clearly established control of the Bakhmut train station by April 20?" - the "clearly established" part helps reflect how ISW does their mapmaking.
But I suspect people would prefer to predict a question that doesn't hinge on the "clearly established" part (which could be easily done by using the orange shaded region, or using a different map). Then the question is more like, "Will Russia appear to have control of the train station?" But do note that such a question could also have come down to the wire, and seen lots of disagreements between different maps, if the deadline was about a week earlier.
I guess one solution could be to resolve "yes" if ISW shades something red and N/A if ISW shades something orange.
That's an option but I'm not a fan of that, I'd rather just ask whether ISW will shade the location either red or orange.
@jack I was listening to Ruslan Leviev today, the head of Conflict Intelligence Team, he mentioned that he switched from one map service (I forgot which one, not ISW) to deepstate (as many viewers and Ukrainians would ask him). So perhaps deepstate would be the best one?
Perhaps an alternative for "orange vs red" on the map could be something like "Deepstate shows an area / a point shaded red during each updated for a period of continuous 48 hours". This allows for any corrections to happen and for (dis)confirmations to happen.
@jack Forgot to mention, Ruslan Leviev was also skeptical of some of the Deepstate updates on the map and wasn't entirely certain in their accuracy.
@42irrationalist Yeah, I've used criteria requiring the map to show Russian control for some contiguous period of time on other questions, like in https://manifold.markets/jack/will-ukraine-control-kherson-city-c. It has pros and cons, makes the question hinge less on map updates that might be based on not much data, but on the other hand it can be a bit of a pain to check whether it hit the resolution criteria or not when there's really a lot of back-and-forth fighting.
@jack AFAIK deepstate allows one to view the previous versions, so it shouldn't be that difficult for singular points. With ISW maps it's much harder I think.
@DavidKochanov btw here is a list of western and ukrainian maps showing the station as russia controlled: https://deepstatemap.live/#15/48.6034/37.9848 https://newsmap.pl/2022-russian-invasion-of-ukraine-interactive-map/ https://liveuamap.com/ https://geoconfirmed.org/ukraine https://www.google.com/maps/d/viewer?mid=1V8NzjQkzMOhpuLhkktbiKgodOQ27X6IV&ll=47.9443760472777%2C35.18634928482012&z=6 https://twitter.com/NOELreports/status/1648950801979588609?s=20 there is some others but as far as i can tell ISW is the only one that doesn't
@higherLEVELING do you mean if ISW update by 3pm? I think they already updated the map but haven't changed the date yet.
@Blomfilter That's funny :) But if Russian forces claim to have full control over the station and want this to be considered confirmed - well, they could post a video from there.