I have read that the answer scores are calculated by the formula (Upvotes - Downvotes). I have just done a query using Stack Exchange Data Explorer (SEDE) and see a discrepancy between this formula and the first (most voted, accepted) answer, while all other answers seem to agree with the formula.

See for yourself with the example below. The first answer (most voted, accepted) has a score of 1380, but it should have 1377, since there are 1378 upvotes and 1 downvote, according to the query.

Here is the post I was testing: What are the correct version numbers for C#?

And here is the query I did: http://data.stackexchange.com/stackoverflow/query/425977/answer-up-down-votes-from-a-specific-question

  • The query actually shows 1 downvote; the additional upvotes are new since the time the data was last updated; the split is 1381/-1 now – Tim Stone Jan 20 '16 at 23:47
  • @TimStone thank you! I mistakenly wrote down 2 downvotes while the right number should have been 1 (corrected). But the discrepancy remains and it was explained by your duplicate post link about the SEDE update policy. Thanks for for clarifying the not-realtime nature of data on Data Explorer, I mistakenly assumed it would be synced with production data. – Rabbit Jan 21 '16 at 23:37
  • @TimStone I am new to SE/SO so that why I am asking. Would it be a good move to write down an answer for my question even it being a duplicate? – Rabbit Jan 21 '16 at 23:39

Browse other questions tagged .