I found this article, might be of help.
One area of football betting that has seen a surge in popularity over the past few years is “halftime wagering” where a player can place a stake on the outcome of a “half” rather than a full game.
The lines for the 1st half of games are a simple enough prospect, and are often close to the full line on a game divided by two. There’s usually a slight penalty to the favorite (so a –7 line might be –4.5 for the favorite in the first half), but by and large you can predict the lines with some accuracy if you know the full spread. The sportsbooks can set the 1st half numbers well in advance of the game, and suffer no more than the typical exposure in the event of a bad line.
On the other hand, the second half lines that get set in the short time between halves are where the real adventure takes place, for the linemakers, the books, and the players who elect to jump in!
For these latter wagers, linemakers send over a hastily created second half line and the casinos book a flurry of bets in the few minutes before the game resumes. Clearly when posting a number with so little time to assess everything there is substantial room for error, and this is one reason why the bets are finding an audience among sharp players.
Now unfortunately we don’t have a database built that contains halftime lines, so the following research is based on using the full line whole or divided by two to reflect an approximation of a halftime line. Again for the 1st half of games this will be reasonably close, while the second half may in some cases be far off.
Nonetheless, we should be able to glean some patterns to how teams respond to events in the early stages of the game. Our first test is how often the team with the lead in the first half (against half the full spread or straight-up; we’ll look at it both ways) gets the “cover” in the second half against the full line divided by two. In other words, if we treat the second half spread as half the full line (which is an approximation) how often does the team ahead cover in the second half?
NFL Half Comparisons 1983-2002: % of the time the first half leader “covers” a second half line of the full spread divided by two
Team Type 1st Half "Spread" Leader 1st Half score Leader
ALL 45% 43%
Home Favorites 41% 41%
Home Underdogs 50% 48%
Away Favorites 44% 43%
Away Underdogs 49% 46%
So in general, the team that fared well in the first half is unlikely to cover a second half line of “half the full spread” – this is particularly true for favorites. In terms of the actual second half lines, what often happens is that an underdog that’s ahead will be receiving a line that is closer to the full value spread than half of it. Likewise a big favorite that’s far in the lead will not be facing half of the full spread again at that point.
The numbers in the table can also be flipped around to learn the comparable percentage for the team that is behind. For instance an away underdog that’s behind in the first half will “cover” half the full line in the second half at a healthy 59% clip.
The value then seems to be in going against favorites with a lead who are giving at least half the original spread in points in the second half (better of course if they are giving more!). Meanwhile underdogs with the lead can be solid plays if they are getting more than half the original spread in the second half line. There are key numbers in the NFL for spreads (eg 3’s and 7’s) that make it elaborate to give a hard and fast rule as to what the difference in an extra point translates into for a winning percentage. Rest assured though that one or two extra points can usually boost your expected winning percentage significantly!
A more detailed breakout of the above follows, showing the relative lead of the team in the first half --
1st Half "SPREAD" lead, 2nd half "line" performance 1st Half Lead 1 - 2 3 - 6 7 - 9 10 - 13 14+ Pts ALL
HomeFavs 154 - 191 167 - 237 115 - 175 105 - 178 138 - 201 679 - 982
HomeDogs 100 - 88 108 - 95 63 - 77 55 - 65 79 - 83 405 - 408
AwayFavs 55 - 83 110 - 116 67 - 63 42 - 70 57 - 82 331 - 414
AwayDogs 189 - 169 238 - 186 124 - 154 90 - 121 99 - 152 740 - 782
Favorites 209 - 274 277 - 353 182 - 238 147 - 248 195 - 283 1010 - 1396
Underdogs 289 - 257 346 - 281 187 - 231 145 - 186 178 - 235 1145 - 1190
Home Teams 254 - 279 275 - 332 178 - 252 160 - 243 217 - 284 1084 - 1390
Away Teams 244 - 252 348 - 302 191 - 217 132 - 191 156 - 234 1071 - 1196
ALL PICKS 498 - 531 623 - 634 369 - 469 292 - 434 373 - 518 2155 - 2586
WIN % 48 % 50 % 44 % 40 % 42 % 45 %
Spread Range 10+ points 5 to 9.5 0 to 4.5
Favorites 131 - 195 387 - 543 492 - 658
Underdogs 148 - 131 391 - 410 606 - 649
So the bigger the "spread" lead, the less likely it is the team will cover a second half line of the full line divided by two. This is especially true of favorites that grab a lead in the first half of more than half the full spread on the game by 10+ points, as these favorites are just 342-531 (39%) against a second half line that is half the full line. Now it may well be that when a favorite grabs a huge lead the second half spread is adjusted down in ratio from the full one, but in the event that it is not, you now know where the value is!
Let's complete the look here by taking the same table from a "straight-up" first half perspective (eg the team that's in the lead on the scoreboard):
1st Half SCORE lead, 2nd half "line" performance 1st Half Lead 1 - 2 3 - 6 7 - 9 10 - 13 14+ Pts ALL
HomeFavs 30 - 50 200 - 264 134 - 171 152 - 239 251 - 388 767 - 1112
HomeDogs 16 - 29 108 - 95 53 - 70 47 - 60 68 - 63 292 - 317
AwayFavs 19 - 32 81 - 127 77 - 90 87 - 84 91 - 139 355 - 472
AwayDogs 47 - 35 190 - 183 88 - 118 68 - 114 72 - 101 465 - 551
Favorites 49 - 82 281 - 391 211 - 261 239 - 323 342 - 527 1122 - 1584
Underdogs 63 - 64 298 - 278 141 - 188 115 - 174 140 - 164 757 - 868
Home Teams 46 - 79 308 - 359 187 - 241 199 - 299 319 - 451 1059 - 1429
Away Teams 66 - 67 271 - 310 165 - 208 155 - 198 163 - 240 820 - 1023
ALL PICKS 112 - 146 579 - 669 352 - 449 354 - 497 482 - 691 1879 - 2452
WIN % 43 % 46 % 44 % 42 % 41 % 43 %
Spread Range 10+ points 5 to 9.5 0 to 4.5
Favorites 179 - 261 428 - 644 515 - 679
Underdogs 66 - 68 215 - 307 476 - 493
More of the same kind of confirmation of the basic tenets to take away -- go against favorites that had the upper hand in the first half if they are laying at least half the original spread in the second half.
An alternative way of looking at things is asking “how often does a team with a first half lead simply outscore its opponent in the second half?”
1st Half SCORE lead, Second Half SCORE results 1st Half Lead 1 - 2 3 - 6 7 - 9 10 - 13 14+ Pts ALL
HomeFavs 46 - 33 235 - 188 156 - 116 197 - 169 309 - 262 943 - 768
HomeDogs 11 - 31 80 - 104 40 - 71 38 - 65 52 - 67 221 - 338
AwayFavs 20 - 26 100 - 79 88 - 62 92 - 65 110 - 98 410 - 330
AwayDogs 30 - 43 123 - 220 65 - 134 49 - 126 48 - 112 315 - 635
Favorites 66 - 59 335 - 267 244 - 178 289 - 234 419 - 360 1353 - 1098
Underdogs 41 - 74 203 - 324 105 - 205 87 - 191 100 - 179 536 - 973
Home Teams 57 - 64 315 - 292 196 - 187 235 - 234 361 - 329 1164 - 1106
Away Teams 50 - 69 223 - 299 153 - 196 141 - 191 158 - 210 725 - 965
ALL PICKS 107 - 133 538 - 591 349 - 383 376 - 425 519 - 539 1889 - 2071
WIN % 45 % 48 % 48 % 47 % 49 % 48 %
Spread Range 10+ points 5 to 9.5 0 to 4.5
Favorites 275 - 129 545 - 424 533 - 545
Underdogs 22 - 100 144 - 349 370 - 524
So it turns out that favorites with a lead do outscore opponents in the second half as well most of the time, but not to the degree as we discovered earlier that they will cover the second half spread (presuming it’s half the full spread or more). Dogs likewise with a lead seldom outplay the opposition in the second half – but again unless you are looking at second half “money-lines” the first table gives you a better sense of which teams to back against a second half spread.