PutFinder › Methodology› 方法论
How PutFinder scores a cash-secured putPutFinder 如何为现金担保看跌期权评分
PutFinder documents every factor behind the score and how they combine into a single relative rank. This page describes the shape and philosophy of that model. PutFinder 记录了分数背后的每一个因子，以及它们如何组合成一个相对排名。本页描述该模型的结构与理念。
What the score is — and what it is not这个分数是什么——不是什么
The PutFinder score is a relative ranking heuristic, not a return predictor, a valuation, or a recommendation. A high score means a contract sits near the top of today's pool on a premium-versus-risk basis — it does not say the trade is “good,” and it does not forecast a profit. Annualized yield is arithmetic on the current premium, not an expected return. Probability of profit (PoP) is a model estimate from the option's delta, not a historical win rate. The decision always stays with you; PutFinder ranks and documents, it does not advise. PutFinder 分数是一个相对排名启发式，不是收益预测、估值或建议。高分表示某合约在今天的池中、以权利金对风险为基础排名靠前——它不代表这笔交易“好”，也不预测盈利。年化收益率是对当前权利金的算术换算，并非预期收益。盈利概率（PoP）是基于期权 Delta 的模型估算，而非历史胜率。决策权始终在你手中；PutFinder 负责排名与记录，而不提供建议。
The factors评分因子
Each eligible contract is measured on a small set of factors. Every one is computed from stored raw inputs and is reproducible — nothing is invented. To compute the arithmetic factors — annualized yield, breakeven, and downside buffer — for a single contract on your own inputs, use the free cash-secured put calculator. 每个符合条件的合约都会在一组因子上被度量。每个因子都由存储的原始输入计算得出、可复现——没有任何数据是凭空捏造的。若要为单一合约、基于你自己的输入计算其中的算术因子——年化收益率、盈亏平衡价与下跌缓冲——可使用免费的现金担保看跌期权计算器。
Stock Score标的分数
- Underlying quality标的质量
- A curated read on the underlying stock, so the premium is weighed against the kind of name you would be assigned. Screening is limited to a curated universe of liquid US stocks.对标的股票的精选评估，使权利金与你可能被行权买入的标的类型相权衡。筛选仅限于精选的美股流动性股票池。
- Volatility risk premium (VRP)波动率风险溢价（VRP）
- The gap between a stock's implied and realized volatility — the structural edge option sellers are compensated for. Computed from one canonical IV30 measure, not recomputed inline.股票隐含波动率与已实现波动率之间的差值——期权卖方因承担风险而获得补偿的结构性来源。由统一的 IV30 口径计算，而非逐处重算。
- Trend趋势
- Rewards a long-term uptrend sitting near, but not below, a support level (its 50-day average), and penalizes a falling knife. A stock-level posture, separate from the strike-level support check in downside buffer below.奖励长期处于上升趋势、且靠近（而非跌破）支撑位（50日均线）的标的，并对下跌趋势中的标的扣分。这是标的层面的走势判断，与下方下跌缓冲中行权价层面的支撑检查是独立的两件事。
Contract Fit合约契合度
- Annualized yield年化收益率
- The option premium expressed as a per-year rate on the capital secured. A comparison figure — arithmetic on the current premium, not an expected or guaranteed return.将期权权利金按占用资金折算成年化利率。它是一个用于比较的数字——对当前权利金的算术换算，并非预期或保证收益。
- Downside buffer下跌缓冲
- How far the stock can fall, as a percentage of the current price, before it reaches the strike. A bigger buffer means more room to be wrong. Buffer is measured to the strike, not to the (lower) breakeven.股价跌至行权价前的空间，以现价的百分比表示。缓冲越大，容错空间越大。缓冲以行权价为界，而非更低一些的盈亏平衡价。
Catalyst multiplier催化剂乘数
- Earnings gate财报门控
- Earnings that land before an option expires add event risk. Applied as a multiplier (0.0–1.0) on the final score — not a weighted term — so a severe gap can zero out an otherwise high-scoring setup rather than merely discounting it.在期权到期前公布的财报会带来事件风险。以乘数（0.0–1.0）的方式作用于最终分数，而非加权项——因此严重的风险时点可以将原本高分的合约直接归零，而不仅仅是打折扣。
Displayed alongside the score — not a scoring input与分数一并展示 — 并非评分输入项
- Probability of profit (PoP)盈利概率（PoP）
- A model-estimated probability the option expires worthless, derived from its delta. Shown on every card as context, not as a factor the score is weighted on. A model estimate, not a historical win rate.基于 Delta 的模型估算——期权到期作废的概率。它作为参考信息展示在每张卡片上，而不是分数所加权的因子。这是模型估算值，不是历史胜率。
Eligibility gate — applied before scoring资格门槛 — 评分前应用
- Liquidity流动性
- Whether the contract is tradable in practice — tight enough markets and real volume — rather than a quote that only looks attractive on screen. A contract that fails this gate is discarded before scoring and never appears; it does not lower the score.合约在实际中是否可交易——买卖价差足够窄、有真实成交量——而不是一个只在屏幕上看着诱人的报价。未通过此门槛的合约会在评分前被剔除、从不出现；它不是降低分数，而是直接不参与评分。
How the factors combine因子如何组合
The factors combine in two layers, then a gate: Final Score = (Stock Score ÷ 100) × Contract Fit × the earnings-gate multiplier. This is deliberately multiplicative, not additive — a great stock with a weak contract, or a strong contract on a falling-knife stock, both score low, since either layer can pull the product down. Probability of profit is estimated separately and shown alongside the score as context; liquidity is an eligibility gate applied before any of this scoring runs. The weightings and the bands that map each factor onto the score are calibrated over time in configuration — not fixed published constants — so the model can be recalibrated against the data without rewriting the logic. Ties are broken deterministically, so the same inputs always produce the same ordering. The output is a shortlist ranked by that relative score, with every input shown next to it. 这些因子分两层组合，再经过一道门控：最终分数 =（标的分数 ÷ 100）× 合约契合度 × 财报门控乘数。这是刻意设计为乘积关系，而非相加——一只好股票配上一个契合度弱的合约，或者一个契合度很强的合约配上一只正在下跌的股票，都会得到低分，因为任何一层都能把乘积拉低。盈利概率是单独估算的，作为参考信息与分数一并展示；流动性是在整个评分开始之前应用的资格门槛。将每个因子映射到分数上的权重与区间是在配置中随时间校准的——而非固定公布的常数——因此模型可以针对数据重新校准，而无需改写逻辑。并列情况以确定性方式打破，因此相同的输入总是产生相同的排序。输出是一份按该相对分数排名的候选清单，每个输入都并列显示。
Eligibility and the data合约资格与数据
Before scoring, contracts must clear eligibility: a 14–45 day time-to-expiration window (centered around one month), the liquidity filters above, a place in the curated universe, and fresh data — stale quotes are excluded rather than scored on old numbers. The option chain comes from a real-time data provider and is treated as the source of truth: prices, implied volatility, greeks, and volume are never fabricated. All option prices are handled on a strict per-share basis, and where a chain delta is missing, a documented Black–Scholes fallback is used and the source recorded. Because every derived metric traces back to stored raw inputs, a rank is reproducible. 在评分之前，合约必须通过资格筛选：14–45 天的到期时间窗口（以大约一个月为中心）、上述流动性过滤、位于精选股票池之内，以及数据新鲜——过期报价会被排除，而不是用旧数据评分。期权链来自实时数据提供方，并被视为事实来源：价格、隐含波动率、希腊字母和成交量绝不捏造。所有期权价格严格按每股口径处理；当期权链缺少 Delta 时，使用有记录的 Black–Scholes 回退并记录来源。由于每个派生指标都可追溯到存储的原始输入，排名是可复现的。
Why these factors — the research为什么是这些因子——相关研究
Selling cash-secured puts is a form of harvesting the volatility risk premium: index-level put-writing has a long, published track record, and the gap between implied and realized volatility is a documented, studied phenomenon. PutFinder does not claim to reproduce those results; the references below describe the structural ideas the factors above lean on. The same premium-selling structure sits on the opposite side of a position in a covered call — see cash-secured put vs covered call. 卖出现金担保看跌期权是一种获取波动率风险溢价的方式：指数层面的卖出看跌期权拥有长期、公开的历史记录，隐含波动率与已实现波动率之间的差距是一个有据可查、被广泛研究的现象。PutFinder 并不声称能复现这些结果；下方参考文献描述了上述因子所依赖的结构性理念。同样的卖出权利金结构也存在于持仓另一侧的备兑看涨期权中——参见现金担保看跌期权 vs 备兑看涨期权。
- Cboe S&P 500 PutWrite Index (PUT) — a benchmark for systematically selling cash-secured puts: cboe.com/us/indices/dashboard/PUT.Cboe 标普500 PutWrite 指数（PUT）——系统性卖出现金担保看跌期权的基准：cboe.com/us/indices/dashboard/PUT。
- The Options Industry Council (OIC) — options basics, assignment, and the cash-secured put: optionseducation.org.期权行业协会（OIC）——期权基础、行权指派与现金担保看跌期权：optionseducation.org。
- Carr, P. & Wu, L. (2009). “Variance Risk Premiums.” The Review of Financial Studies 22(3): doi.org/10.1093/rfs/hhn038.Carr, P. 与 Wu, L.（2009）。“Variance Risk Premiums.”《The Review of Financial Studies》22(3)：doi.org/10.1093/rfs/hhn038。
How it compares与其他方式的对比
A descriptive comparison of ways to find cash-secured-put setups — not a claim that one is “better,” just how they differ. 对寻找现金担保看跌期权合约的几种方式的描述性对比——并非主张某一种“更好”，只是说明它们的区别。
| PutFinder | Manual chain-scanning手动翻期权链 | Signals service / Discord交易信号服务 / Discord | |
|---|---|---|---|
| What you get你得到什么 | A ranked, documented shortlist带说明的排序候选清单 | Raw data you sort yourself需自己筛选的原始数据 | Specific calls to act on具体的操作提示 |
| Transparency透明度 | Every factor documented每个因子都有说明 | Your own method取决于你自己的方法 | Usually opaque通常不透明 |
| Coverage覆盖范围 | Full chain, curated universe, daily完整期权链、精选股票池、每日更新 | Whatever you have time to scan取决于你能扫描多少 | A handful of tickers少数几个标的 |
| Framing定位 | Descriptive — you decide描述性——由你决定 | —— | Prescriptive — told what to do指令性——直接告诉你怎么做 |
| Not investment advice是否为投资建议 | ✓ research tool✓ 研究工具 | ✓ your own research✓ 你自己的研究 | Varies — often implies action不一定——常暗示操作 |
Common questions常见问题
Is the score a prediction of return?这个分数是对收益的预测吗？
No. The score is a relative ranking heuristic, not a return predictor, valuation, or recommendation. It says a contract sits high in today's pool on a premium-vs-risk basis — it does not forecast a profit. Annualized yield is arithmetic on the current premium, and PoP is a model estimate from delta, not a historical win rate.不是。该分数是相对排名启发式，不是收益预测、估值或建议。它表示某合约在今天的池中、以权利金对风险为基础排名靠前——并不预测盈利。年化收益率是对当前权利金的算术换算，PoP 是基于 Delta 的模型估算，而非历史胜率。
Does PutFinder publish its exact weights and thresholds?PutFinder 会公布确切的权重和阈值吗？
PutFinder documents every factor and how the factors combine into the score, with the weightings and bands tuned over time against the data. It is transparent about the shape and philosophy of the model; the exact live cutoffs live in configuration and are recalibrated, not published as fixed numbers.PutFinder 记录每一个因子以及因子如何组合成分数，权重与区间会随时间针对数据调校。它对模型的结构与理念保持透明；确切的实时阈值存放在配置中并会重新校准，而不作为固定数字公布。
Does PutFinder tell me what to sell?PutFinder 会告诉我该卖什么吗？
No. PutFinder ranks and documents contracts on a premium-vs-risk basis; the decision stays with you.不会。PutFinder 以权利金对风险为基础对合约进行排名并记录；决策权始终在你手中。
Last updated . 最后更新 。