当特朗普总统知道自己陷入麻烦时，你是能看出来的。当然，他永远不会承认自己犯了错。如果说有哪件事到现在人人都该明白，那就是特朗普永远不会承认自己错了，永远不会承认自己输了，也永远不会道歉。
Trump’s midterm reckoning: Why the Republican Party is facing a Senate wipeout You can tell when President Trump knows he is in trouble. Of course, he’ll never admit to making a mistake. If there is something everyone should know by now, it’s that Trump will never admit he is wrong, never admit he lost and never apologizes.
但在过去几周里，特朗普已明确释放出信号，表明他认识到自己对通胀问题重视不够——与加拿大的贸易争端至少可以说时机不当——而且他挑了几个实在不怎么样的人去竞选参议院席位。
But over the past few weeks, Trump has clearly signaled he recognizes that he has not taken inflation seriously enough — the trade fight with Canada was ill-timed, to say the least — and he picked some real duds to run for Senate.
尽管他认识到自己的错误值得肯定——这与拜登政府的否认和自欺形成了鲜明对比——但纠正错误为时已晚。尽管各种偏保守的评论人士对糟糕的民调和假新闻大肆兜售自我安慰的说辞，而且参议院的选举版图实际上对共和党有利，但共和党失去参议院控制权的可能性仍然大于保住它，此外还会丢掉众议院多数席位。
Although he deserves some credit for recognizing his errors, in marked contrast to the Biden administration’s denial and delusion, it is too late to fix his mistakes. Despite the copious copium dished out by various conservative-leaning pundits over bad polls and fake news, and what is really a favorable Senate map, Republicans are more likely than not to lose control of the Senate in addition to their majority in the House.
一年前，共和党失去参议院控制权看起来还相当不可能。在特朗普2024年落败的州中，唯一需要争夺的席位在缅因州，而参议员苏珊·柯林斯（共和党）已被证明是终极的政治幸存者。与此同时，民主党要在特朗普赢下的两个州——佐治亚州和密歇根州——捍卫席位。即便特朗普和共和党面临典型的执政首任期中期选举颓势周期，失去多数席位似乎也是不可想象的。
A year ago, losing control of the Senate seemed rather unlikely for Republicans. The only seat in a state where Trump lost in 2024 was Maine, where Sen. Susan Collins (R) has proven to be the ultimate political survivor. At the same time, Democrats were defending seats in two states Trump won — Georgia and Michigan. Even if Trump and the Republicans faced a typical first-term downer cycle, losing the majority seemed inconceivable.
如果每个州的每位共和党人都比2024年特朗普的表现落后10个百分点，共和党也只会丢掉缅因州和北卡罗来纳州。这是巨大的结构性优势。在目前被视为竞争激烈的席位中，特朗普的胜选优势从11.2个百分点（俄亥俄州）到17.9个百分点（南卡罗来纳州）不等。然而根据RealClearPolitics的平均数据，民主党候选人在所有2024年竞争激烈的州和民主党州，以及俄亥俄州、阿拉斯加州（特朗普领先13.1个百分点）和得克萨斯州（特朗普领先13.7个百分点）都处于领先。
If every Republican in every state ran 10 points behind 2024 Trump, Republicans would lose only Maine and North Carolina. That is a monumental structural advantage. Trump’s winning margin in the seats that are now viewed as competitive ranged from 11.2 percent (Ohio) to 17.9 percent (South Carolina). Yet according to the RealClearPolitics averages, Democratic candidates are winning in all the 2024 competitive and Democratic states, as well as in Ohio, Alaska (plus-13.1 percent for Trump) and Texas (plus-13.7 percent for Trump).
参议院最终席位分布可能在共和党 53 比 47 占优（现状不变）到民主党 55 比 45 大胜之间，竞争激烈的席位足以导致这种巨大跨度。目前形势越来越倾向于民主党至少获得微弱多数。撰写本文时，RealClearPolitics “无摇摆州”版地图显示民主党 52 比 48 领先；加入 7 个摇摆州后为 47 比 44；Sabato’s Crystal Ball 显示 49 比 46，有 5 个摇摆州；Cook Political Report 显示 47 比 46，有 7 个摇摆州。
There are enough close races that the final tally for the Senate could be anywhere from a 53-47 Republican majority (no change) to a 55-45 Democratic wipeout. It is increasingly looking like at least a slim Democratic majority. At the time of writing, the RealClear “no toss-ups” map is 52-48 for Democrats. With their seven toss-ups, RealClear has it at 47-44, Sabato’s Crystal Ball shows 49-46, with five toss-ups, and Cook Politico Report is at 47-46, with seven toss-ups.
预测最终席位的难点在于，竞争性选区的民主党候选人民调支持率均未稳定达到或超过 50%。这意味着未决定选民将决定胜负——这些未决定选民大多是 2024 年特朗普的选民。这一数据支撑了特朗普和共和党领导层的信心。
The challenge in projecting the final numbers is that none of the Democrats in competitive races are consistently polling at or above 50 percent. That means undecided voters will make the difference — most of those undecided voters were 2024 Trump voters. That bit of data is what gives Trump and Republican leaders’ confidence.
然而，这种推理的问题在于：这些未决定选民既必须出来投票，又不能倒向共和党的对手。
The problem with that reasoning, however, is that those undecideds must both turn out and not turn against the Republicans.
以俄亥俄州为例，特朗普以逾 11 个百分点优势获胜。RealClearPolitics 平均民调显示，前参议员谢罗德·布朗（民主党）以 48.1% 对 44.8% 领先现任参议员乔恩·哈斯特德（共和党，系由 J.D. 万斯空缺补任），未决定选民占 7.1%。若采信民调，哈斯特德要反超布朗，需赢得 75% 的未决定选民——前提是他们真的去投票。若仅有一半未决定选民投票，哈斯特德将落败。
Consider Ohio, where Trump won by more than 11 points. Former Sen. Sherrod Brown (D) leads current Sen. Jon Husted (R) — appointed to the JD Vance vacancy — 48.1 to 44.8 percent, with 7.1 percent undecided, in the RealClear average. If you accept the polling, then for Husted to overtake Brown, he would need to win 75 percent of undecided voters — and that’s if they actually vote. If only half the undecided voters turn out, Husted loses.
得克萨斯州形势稍好：民主党州众议员詹姆斯·塔拉里科在 RealClearPolitics 平均民调中仅以 1.8 个百分点领先，未决定选民占 4.8%。共和党的肯·帕克斯顿需获得 71% 的未决定选民支持。若仅有一半未决定选民投票，帕克斯顿将落败。较小的领先优势意味着更多民调处于误差范围内，帕克斯顿实际支持率可能比民调更接近。但误差范围是双向的。
The situation is brighter in Texas, with Democratic state Rep. James Talarico having just a 1.8 percent RealClear advantage and 4.8 percent undecided. Republican Ken Paxton needs 71 percent of the undecided voters. If only half the undecided voters turn out, Paxton loses. The narrower lead does mean that more polls are within the margin of error. Paxton might be closer than reported. But margin of error works both ways.
是的，民调可能有偏差，但这与全国民意不符。认为国家走向“正确方向”与“错误轨道”的比例净值为负31分以上。特朗普在RealClearPolitics平均支持率中净值为负近22分。即使是对特朗普友好的民调也显示总统落后，拉斯穆森报告为负16分，Morning Consult为负17分，RMG Research为负15分。其中最亲特朗普的哈里斯-哈佛民调显示特朗普落后20分。哪怕出现有利于共和党1个百分点的系统性民调误差（这已经很多了），民主党在大多数竞选中仍将领先，共和党只能指望动员不满的选民。
Yes, the polls might be off, but that does not track with the national mood. The right direction-wrong track is net negative by more than 31 points. Trump’s RealClear average approval is negative by almost 22 points. Even Trump-friendly polls show the president lagging, with Rasmussen at negative 16 points, Morning Consult down 17 points and RMG Research down 15 points. The most Trump-friendly of them all, Harvard-Harris, has Trump down 20 points. A systemic polling error of 1 point in the Republicans’ favor (which is a lot), would still leave Democrats leading most races and Republicans relying on turning out sour voters.
共和党有可能守住50或51个席位。阿卜杜勒·埃尔-赛义德在密歇根州是个糟糕的候选人，顽固地维持在约2个百分点的领先优势。《纽约时报》关于玛丽·佩尔托拉阴谋诡计的报道，可能足以助现任参议员丹·沙利文（共和党-阿拉斯加州）一臂之力。特洛伊·杰克逊在缅因州对阵科林斯时是个弱势候选人。
It is possible Republicans will hang on to 50 or 51 seats. Abdul El-Sayed is a bad candidate in Michigan, stubbornly stuck at about a 2-point lead. The New York Times story on Mary Peltola’s skullduggery could be enough to boost the real Sen. Dan Sullivan (R) in Alaska. Troy Jackson is a weak candidate in Maine against Collins.
但他们的参议院多数地位竟然岌岌可危，本身就是对共和党的严厉谴责。一个仅拥有微弱多数的新参议院，仍将有丽莎·穆尔科斯基（共和党-阿拉斯加州）这类不可靠人物、刺头兰德·保罗（共和党-肯塔基州）和变数约翰·费特曼（民主党-宾夕法尼亚州）。再加上一位或许并不秘密地认为自己应能再次竞选、且肯定不愿放弃任何权力的总统。无论周二结果如何，未来几年都将喧嚣且充满派系斗争。
But the fact that their Senate majority is in doubt at all is a major rebuke to Republicans. A new Senate with a bare Republican majority would still feature unreliables like Lisa Murkowski (R-Alaska), the prickly Rand Paul (R-Ky.) and the wildcard John Fetterman (D-Pa.). And all that with a president who maybe-not-so-secretly thinks he should be able to run again and certainly is not interested in relinquishing any power. Regardless of the results on Tuesday, it will be a raucous, fractious couple of years.
基思·诺顿曾任竞选顾问，是公共事务公司“沉默多数策略”的联合创始人。可在其Substack专栏“押注政治”中找到他。
Keith Naughton is a former campaign consultant and co-founder of the public affairs firm Silent Majority Strategies. Find him on his Substack, “Betting On Politics.”