A closing total is the predicted combined score for both teams after wagers have been placed and the betting line has closed. Also called the closing over/under, it is the number bettors use to choose whether the actual combined score will finish over or under the line; the winner and margin do not matter. section 17.0 section 17.3
The total can move from its opening value as betting arrives. A move upward is intended to encourage under wagers, while a move downward is intended to encourage over wagers. For Memphis–Indiana, the total moved from 209 to 204.5, indicating heavier betting on the under. section 17.0
To compare a closing total with the opening total and the game result, create these game-level measures:
- Add road and home points to produce the actual combined score, ptsT. section 17.4.2
- Compute each line’s absolute error: abs(ptsT - opentotal) and abs(ptsT - closetotal). section 17.4.2
- Classify movement as up, down, or same, then label which total has the smaller error. section 17.4.2
2018–19 NBA comparison
Across the 2018–19 regular season, closing totals were closer to combined points more often than opening totals, by about 10% in the count comparison. Their average variance from combined points was about 2% lower, and the count advantage held whether the total had moved up or down. section 17.5.1 section 17.5
oddsdf3 %>%
mutate(versusPTS =
case_when(diff_ptsT_opentotal >
diff_ptsT_closetotal ~ "closetotal",
diff_ptsT_closetotal >
diff_ptsT_opentotal ~ "opentotal",
diff_ptsT_opentotal ==
diff_ptsT_closetotal ~ "same")) -> oddsdf3
oddsdf3$versusPTS <- factor(oddsdf3$versusPTS)
This R transformation assigns each game to closetotal, opentotal, or same according to which line is nearer the final combined score. section 17.4.2
These comparisons support a collective-intelligence interpretation: bettors collectively added information beyond the opening odds. The finding is limited to the examined NBA regular season, rather than establishing that every closing total will be more accurate. section 17.5 section 17.6
