https://338canada.com/record.htm

The 338Canada / Qc125 model has thus far covered 18 general elections in Canada. In total, 2,039 electoral districts were projected.

So far, the model has correctly identified the winner in 1,821 districts, a success rate of 89.3%.

Among the 218 remaining districts, 132 of the winners (6.5%) obtained a share of the vote that was within the projection's margin of error (moe). Only 86 districts (4.2%) were complete misses.

On this site, electoral districts are labelled according to the odds of winning by the projected favourite party. If a party's odds of winning are at or above 99.9% (formerly 99.5%), the district is labelled as safe. Between 90% and 99.9%, the district is likely. Between 70% ad 90%, we use the label leaning. Finally, if no party has odds above 70%, the district will be labelled as a toss up.

Here below is the complete breakdown of correct and incorrect winners from safe, likely, leaning and toss up districts:

It's a bad attempt at copying Nate Silver's work, but lacks any statistical quality or funding. It's something to read but in my opinion, total BS.