ALCS Game 2, four hockey games chosen for their goalies, and the bottom half of the last sixteen in Shanghai overnight. In hockey tonight the ratings and the odds agree on every game we looked at, so the goalies are the story: who is saving more than he should, and which teams are playing better than their record. How to read this: for each game we work out what the chances should be, then compare that with what the betting odds say.
Parker Messick's expected ERA (what it should be, judged by how hard hitters actually hit him) is 3.15; Freddy Peralta's is 3.84. That is about two-thirds of a run in Cleveland's favour, and Messick held Tampa to two runs in 13⅓ innings across two starts this season.
The odds still make Tampa 55%, and have firmed a little since the open. That is the price of home field and a bullpen that had five days off before Game 1.
Against it Messick has walked six in 6⅔ playoff innings and did not get through the fifth in either Division Series start. Peralta has been better than his season line: a 3.25 ERA since Tampa traded for him and a 0.96 ERA in September. Cleveland's best relievers, Gaddis and Cade Smith, have now pitched three times in four days.
| Starter | IP | ERA | xERA | ERA − xERA | K% | BB% | Barrel% | Hard-hit% | Exit velo |
|---|---|---|---|---|---|---|---|---|---|
| Parker MessickCLE · LHP | 186.1 | 2.56 | 3.15 | −0.59 | 25.8 | 6.6 | 5.0 | 34.3 | 87.6 |
| Freddy PeraltaTB · RHP | 169.0 | 4.42 | 3.84 | +0.58 | 21.4 | 8.6 | 5.8 | 37.2 | 87.6 |
| Messick | IP | H | ER | BB | K | Pitches | Fastball | Whiff % | |
|---|---|---|---|---|---|---|---|---|---|
| Oct 8 | at CWS | 2.1 | 2 | 2 | 4 | 5 | 65 | 95.6 | 41 |
| Oct 3 | vs CWS | 4.1 | 4 | 2 | 2 | 6 | 87 | 96.3 | 33 |
| Sep 27 | at KC | 3.0 | 6 | 1 | 1 | 2 | 64 | 93.7 | 17 |
| Sep 22 | at BOS | 5.0 | 5 | 1 | 1 | 5 | 96 | 94.8 | 17 |
| Sep 16 | vs CWS | 6.0 | 7 | 3 | 0 | 8 | 86 | 95.6 | 30 |
| Peralta | IP | H | ER | BB | K | Pitches | Fastball | Whiff % | |
|---|---|---|---|---|---|---|---|---|---|
| Oct 5 | vs NYY | 4.2 | 3 | 1 | 1 | 3 | 66 | 96.8 | 32 |
| Sep 25 | at PHI | 5.1 | 4 | 0 | 3 | 4 | 93 | 96.2 | 24 |
| Sep 19 | vs BOS | 5.2 | 4 | 1 | 2 | 6 | 97 | 96.4 | 31 |
| Sep 13 | vs HOU | 5.0 | 0 | 0 | 3 | 4 | 86 | 94.6 | 28 |
| Sep 8 | at ATL | 6.0 | 5 | 1 | 1 | 2 | 92 | 94.4 | 16 |
Both favourites lost. Cleveland won ALCS Game 1 7–3: Griffin lasted four innings as expected, and Rasmussen gave up two home runs in six. Milwaukee won NLCS Game 2 7–4 and lead 2–0; Yamamoto, who had gone six innings or more in seven straight starts, was out after 2⅔. Neither card claimed an edge against the price.
The odds give Montreal 58% and the ratings 58%: no disagreement at all. The case for Buffalo is that they have had 56% of the expected goals and are still six goals worse off than their opponents, mostly because Ukko-Pekka Luukkonen has let in about four more than an average goalie.
Montreal's Jacob Fowler, in for the injured Samuel Montembeault, stopped all 30 shots against Detroit on Saturday.
Against it Buffalo played yesterday afternoon and Luukkonen came on in relief for 34 minutes, so he is not confirmed and someone else may start. Their record is 3–3, which is not much of a slump. Fowler has two games this season; the shutout is half of them.
| Odds say | Montreal 58% · Buffalo 42% |
| Ratings say | Montreal 58% at home · under a point apart |
| Fowler, MTL · confirmed | Sat v Detroit: 30 saves on 30; chances worth 3.6, so he saved three and a half more than an average goalie · Thu v Nashville: 15 on 18, a goal worse than average · 12 Apr (last season) at the Islanders: 32 on 33, about three better · Season: .938, two and a half better than average in two games |
| Luukkonen, BUF · <b>unconfirmed</b> | Thu v Dallas: 14 saves on 18; chances worth 2.1, so two more than an average goalie let in · Sat 3 Oct v Chicago: 12 on 15, half a goal worse · Thu 1 Oct at Columbus: 17 on 22, nearly three worse · in relief yesterday: 13 on 14 · Season: .812, about four worse than average |
| The teams | Buffalo 3–3, 56% of expected goals, goal difference −6 against +2 expected · Montreal 2–2–1, 48% |
The odds give Dallas 52% and the ratings give them 52%, so neither has much to say. The goalies do: Jake Oettinger has let in one goal in his last three games, and Mackenzie Blackwood let in five on 22 shots on Saturday.
Over the season so far Oettinger has saved about six and a half goals more than an average goalie would have; Blackwood has let in about four more.
Against it Four games each. Colorado have been the better team in front of their goalie: 58% of the expected goals and a goal difference of +9, while Dallas's +4 is already better than their chances deserve. Last season Blackwood was well above average and Oettinger only slightly, so both are likely to move back towards the middle.
| Odds say | Dallas 52% · Colorado 48% · moved about 2 points towards Dallas |
| Ratings say | Dallas 52% at home · half a point apart |
| Oettinger, DAL · confirmed | Sat at Pittsburgh: 19 saves on 20; the chances were worth 3.1 goals, so he saved two more than an average goalie would have · Thu at Buffalo: 26 on 26, chances worth 2.5 · Mon 5 Oct v San Jose: 13 on 13, chances worth 1.6 · Season: .963, about six and a half goals better than average in four games |
| Blackwood, COL · confirmed | Sat v Toronto: 17 saves on 22; chances worth 2.0, so he let in three more than an average goalie · Wed 7 Oct at Winnipeg: 16 on 19, a goal worse than average · Sat 3 Oct v St Louis: 25 on 26, a goal better · Season: .867, about four goals worse than average |
| The teams | Colorado 3–2, 58% of expected goals · Dallas 2–2–1, 51% |
The odds give the Islanders 65%, the ratings 67%. Ilya Sorokin has saved about four and a half goals more than an average goalie this season, including a 37-save shutout of New Jersey.
Vancouver have lost four straight and deserved to: they have had only 39% of the expected goals.
Against it Carded for the goalie, not for an edge. Sorokin let in four on 25 shots against Tampa on Saturday, his worst game of the year, and Vancouver's likely starter Leevi Merilainen has been about average this season after a very poor one last year.
| Odds say | Islanders 65% · Vancouver 35% |
| Ratings say | Islanders 67% at home · 2 points apart |
| Sorokin, NYI · confirmed | Sat v Tampa Bay: 21 saves on 25; chances worth 2.5, a goal and a half worse than average · Thu v Chicago: 22 on 23, a goal and a half better · Sat 3 Oct v New Jersey: 37 on 37, chances worth 4.0 · Season: .937, about four and a half better than average |
| Merilainen, VAN · likely | Sat at New Jersey: 31 saves on 34; chances worth 3.3, about average · Sat 3 Oct v Calgary: 23 on 24, a little better · 17 Jan (last season, Ottawa) v Montreal: 13 on 19, nearly three worse · Season: .929 in four games, two in relief |
Both teams have lost three of their last four while getting more of the chances than their opponents. New Jersey have had the best share of shots in the league and are seven goals worse off than their opponents; on chances they should be two better. Detroit are four worse off and should be two better.
The odds and the ratings both call it about even, with Detroit a fraction ahead. The difference is in goal: John Gibson was pulled on Friday after letting in five on 15 shots.
Against it Gibson was one of the better goalies in the league last season, and one bad night is most of his number this year. Jake Allen has been only average himself. Everything here rests on four or five games a side.
| Odds say | Detroit 51% · New Jersey 49% |
| Ratings say | Detroit 53% at home · 1.5 points apart |
| Allen, NJD · confirmed | Sat v Vancouver: 20 saves on 21; chances worth 2.3, a goal better than average · Tue 6 Oct v Utah: 26 on 31, about a goal worse · Thu 1 Oct v Philadelphia: 23 on 25, average · Season: .896, about average |
| Gibson, DET · confirmed | Fri v Seattle: pulled after 34 minutes, 10 saves on 15; chances worth 2.0, so three more than an average goalie let in · Tue 6 Oct v Ottawa: 22 on 25, about average · Sun 4 Oct v Winnipeg: 24 on 27, about average · Season: .862, about two and a half worse than average |
| The teams | New Jersey 2–3, 58% of even-strength expected goals, goal difference −7 against +2 expected · Detroit 1–3–1, 57%, −4 against +2 |
The odds give Tommy Paul 65%. The hard-court ratings give him 71%.
Zizou Bergs beat Carlos Alcaraz in straight sets on Monday, at odds of 9.45, and the ratings were last updated before that. Paul beat him 6–1 6–2 at Indian Wells in March.
Against it The ratings have not seen the best win of Bergs's career, so they are likely too low on him by some amount. Paul's foot ended his 2025 season early.
| Odds say | Paul 65% · Bergs 35% |
| Ratings say | Paul 71% · 5 points apart, on Paul |
| Line | Opened Paul 1.48 / Bergs 2.82 · now 1.48 / 2.88 · barely moved, slightly towards Paul |
| This week | Bergs: Kopriva, Buse 6–0 6–3, Alcaraz 7–5 7–6 · Paul: Vallejo 6–2 6–2, Vacherot 6–4 6–2 |
| Head-to-head | Paul 1–0 · Indian Wells 2026, 6–1 6–2 |
| This season | Paul 42–19 (22–11 on hard courts) · Bergs 29–24 (13–12) |
The odds give Rafael Jodar 63%. The ratings give him 66%.
They have never played. Jodar is 28–11 on hard courts this season; Botic van de Zandschulp beat Arthur Fils in straight sets to get here.
Against it Three points is a small gap. Jodar pulled out of Queen's in June with an abdominal problem. Corrected 15 Oct: an earlier version also mentioned a recurring back and hip issue; that report was about Arthur Fils, not Jodar.
| Odds say | Jodar 63% · Van de Zandschulp 37% |
| Ratings say | Jodar 66% · 3 points apart, on Jodar |
| Line | Opened Jodar 1.55 / Van de Zandschulp 2.61 · now 1.56 / 2.62 · flat |
| This week | Jodar: Tirante, Lehecka 4–6 7–6 6–4 · Van de Zandschulp: Mérida, Michelsen, Fils 7–5 6–4 |
| Head-to-head | never played |
The odds give Félix Auger-Aliassime 57%, the ratings 60%.
He leads their meetings 6–4, and the last one, in Monte Carlo in April, ended when Casper Ruud retired with a calf injury.
Against it A small gap. Ruud is 12–10 on hard courts this season and has looked sharp here, beating Tien in straight sets.
| Odds say | Auger-Aliassime 57% · Ruud 43% |
| Ratings say | Auger-Aliassime 60% · 3 points apart, on Auger-Aliassime |
| Line | Opened Ruud 2.21 / Auger-Aliassime 1.74 · now 2.27 / 1.71 · a point towards Auger-Aliassime |
| This week | Ruud: Bu, Tien 6–4 7–6 · Auger-Aliassime: Ugo Carabelli, Carreño Busta 6–3 4–6 6–1 |
| Head-to-head | Auger-Aliassime 6–4 · Monte Carlo 2026, Ruud retired (calf) |
The odds give Daniil Medvedev 66%, the ratings 64%.
He leads the head-to-head 10–5, but Stefanos Tsitsipas won the most recent, in Doha this year, 6–3 6–4, and beat Fritz from a set down to get here.
Against it Medvedev was defaulted in Beijing for hitting a spectator with a ball; this is his first event since. The bookmakers disagree on which way the money is going.
| Odds say | Medvedev 66% · Tsitsipas 34% |
| Ratings say | Medvedev 64% · 2 points apart, on Tsitsipas |
| Line | Pinnacle: opened Medvedev 1.48 / Tsitsipas 2.83, now 1.46 / 2.95, towards Medvedev · bet365: 1.40 / 3.00 to 1.44 / 2.75, towards Tsitsipas |
| This week | Medvedev: Struff, Svrcina 6–2 6–3 · Tsitsipas: Coppejans, Darderi, Fritz 3–6 7–5 6–3 |
| Head-to-head | Medvedev 10–5 · latest Doha 2026 to Tsitsipas, 6–3 6–4 |
Three for three where the ratings disagreed with the odds. De Minaur beat Khachanov 6–3 6–4 despite the injury that had the market at 52%; Zverev beat Bublik in a final-set tiebreak; Davidovich Fokina came from a set down to beat Hurkacz. Gea beat Mensik, where the ratings and the odds had agreed on Mensik.