This Trading Strategy Loses 88% of the Time — Transcript

Full transcript

1. 0:00Let me ask you a question. Would you
2. 0:01trade a strategy that loses 88% of the
3. 0:04time? Because most traders would look at
4. 0:05that strategy and say, "Absolutely
5. 0:08not." And then give you the finger and
6. 0:10say that's a terrible strategy. But what
7. 0:12if I told you that strategy was
8. 0:13profitable? And not theoretically
9. 0:15profitable. I'm talking about the actual
10. 0:17strategy I showed you yesterday. I
11. 0:19analyzed 776 trades and only won about
12. 0:2212% of the time. That means almost nine
13. 0:24out of 10 trades were losers. And yet it
14. 0:27made money. And I think this exposes one
15. 0:29of the biggest understandings that
16. 0:30traders have about win rate. Because
17. 0:33we're obsessed with being right. 50% win
18. 0:35rate, 60% win rate, 70%. You'll see
19. 0:38people advertising strategies with an 80
20. 0:40or 90% win rate. But without knowing one
21. 0:42specific number, the win rate tells you
22. 0:45basically nothing. That number is your
23. 0:47risk to reward ratio. Let me show you
24. 0:49something really simple. Let's ignore
25. 0:51spreads, commissions, and slippage for a
26. 0:53second just to make it mathematically
27. 0:54clean. Every time I lose $100 and every
28. 0:58time I make $100, that's a one-to-one
29. 1:01risk to reward ratio. I need to win 50%
30. 1:04of my trades just to break even. But
31. 1:06watch what happens when my winners
32. 1:07become larger than my losers. If I have
33. 1:10a one-to-1.5, it's 40%. If I have a
34. 1:12one-to-two, it's 33%. A one-to-three is
35. 1:1625%, a one-to-four is 20%, a one-to-five
36. 1:20is 16%, and a one-to-10 risk to reward
37. 1:23ratio needs 9% to break even. Think
38. 1:27about that last one. With a one-to-10
39. 1:29risk to reward ratio, theoretically I
40. 1:31could lose more than 90% of my trades
41. 1:34and still be break even. Obviously, real
42. 1:36trading has costs, execution
43. 1:38differences, slippage, and all the other
44. 1:40[ __ ] But mathematically, that's how
45. 1:42powerful the relationship between win
46. 1:44rate and risk to reward ratio actually
47. 1:47is. And yesterday's strategy gave us a
48. 1:49real-life example. The average winner
49. 1:51was roughly nine to 10 times larger than
50. 1:54the average loser. So the average
51. 1:55historical break even rate is around 9%
52. 1:57to 10% and the actual win rate was 12%
53. 2:01and that's an edge. Not a high win rate,
54. 2:04but the fact that when the strategy
55. 2:06actually catches some winners, that
56. 2:07winner pays off for a pile of losers.
57. 2:10But, there's another problem here
58. 2:12because when I say a 12% win rate, you
59. 2:14might subconsciously imagine the winners
60. 2:16being evenly distributed. Nine losers,
61. 2:19one winner. Nine losers, one winner.
62. 2:22Nice and orderly. But, that's not how
63. 2:24probability works. And the exact same
64. 2:27misunderstanding happens with a 50% win
65. 2:29rate strategy. If I tell you a strategy
66. 2:31wins 50% of the time, your brain wants
67. 2:33to imagine one winner, one loss. One
68. 2:36winner, one loss. Or maybe two
69. 2:38back-to-back losers and two back-to-back
70. 2:40winners. And that over time everything
71. 2:42stays balanced. But, 50% doesn't mean
72. 2:45that at all. It means that over a
73. 2:46sufficiently large sample, approximately
74. 2:49half of the observations were winners.
75. 2:51The order of those winners and losers
76. 2:53can be absolutely brutal. You can have
77. 2:5510 losers in a row, three winners,
78. 2:57another six losses, five winners, and
79. 3:00another nine losses. And eventually over
80. 3:02hundreds or thousands of trades,
81. 3:04everything averages out towards the
82. 3:06strategy's underlying win rate, the 50%.
83. 3:09And that's where trading stops being a
84. 3:10math problem and becomes a human
85. 3:12problem. Think about flipping a coin.
86. 3:14Heads and tails are each 50%, but nobody
87. 3:16expects heads, tails, heads, tails,
88. 3:18heads, tails. And especially not
89. 3:20indefinitely. You could easily get
90. 3:22several heads in a row. That doesn't
91. 3:24suddenly mean the coin stopped being
92. 3:2650/50. And trading results behave the
93. 3:28same way. Except there's one enormous
94. 3:30difference. Money is attached to the
95. 3:32outcome. So, after enough consecutive
96. 3:34losses, you don't calmly say, "Well,
97. 3:37this is perfectly compatible with the
98. 3:38statistics and variables." You say this
99. 3:41strategy doesn't [ __ ] work anymore.
100. 3:43And then you change it. And that's where
101. 3:44traders destroy themselves. Our real
102. 3:47data showed this exactly in yesterday's
103. 3:50video. Across the unrestricted test, we
104. 3:52found a historical losing streak of
105. 3:54approximately 47 trades. 47. Think about
106. 3:58actually experiencing that in your
107. 4:00trading. Not looking at a spreadsheet
108. 4:02after the fact, you're actually trading
109. 4:04it. Loss number 10, loss number 15, loss
110. 4:07number 20. At some point, you're opening
111. 4:09TradingView and thinking, "Something's
112. 4:11changed." Then trade number 30 is a
113. 4:13loss. Now you're changing settings,
114. 4:16you're adding indicators, you're
115. 4:17adjusting the stop loss. Maybe you stop
116. 4:20trading altogether, and then eventually,
117. 4:22a huge winner appears. Except you don't
118. 4:24take it, because you lost the previous
119. 4:2630 trades convinced that your strategy
120. 4:28was broken. And on a strategy designed
121. 4:30for massive winners, missing that one
122. 4:32trade can destroy the entire
123. 4:35mathematical expectancy. That's the
124. 4:37irony. The trader didn't necessarily
125. 4:39lose because the strategy stopped
126. 4:41working. They lost because they couldn't
127. 4:43emotionally survive the distribution of
128. 4:45returns required for the strategy to
129. 4:47work. And this brings me to something
130. 4:48else that I think traders massively
131. 4:50underestimate: time. Let's say we
132. 4:52genuinely identified a strategy with a
133. 4:54positive expectancy. Not something we
134. 4:56curve fitted yesterday. A strategy we've
135. 4:58tested properly and have reasonable
136. 5:00evidence that it contains an edge. That
137. 5:02does not mean it's going to make money
138. 5:04tomorrow. It doesn't necessarily mean
139. 5:06it'll make money next week. It might not
140. 5:08even make money next month. An edge
141. 5:10reveals itself across a distribution of
142. 5:13trades. The larger the sample becomes,
143. 5:15the more opportunity mathematical
144. 5:17advantage has to express itself. two
145. 5:20traders running the exact same
146. 5:21profitable strategy. Trader A trades it
147. 5:24for 3 months. He gets a horrible
148. 5:26sequence, drawdown, losses, a couple of
149. 5:29winners, more losses. He quits. He says,
150. 5:32"Strategy doesn't work." Trader B trades
151. 5:34the same strategy with appropriate risk,
152. 5:37follows it consistently for 3 years. He
153. 5:40experiences good months, bad months,
154. 5:43winning streaks, losing streaks, periods
155. 5:45where it feels amazing, and periods
156. 5:47where it feels completely broken. But,
157. 5:49assuming the edge actually persists, his
158. 5:52much larger sample gives that expectancy
159. 5:54far more opportunity to emerge. And
160. 5:57that's why saying if it has an edge 3
161. 5:59years guarantees profit would actually
162. 6:01be wrong, cuz there's no guarantees. The
163. 6:04better way to say it is, the longer you
164. 6:06execute a genuine positive expectancy
165. 6:09strategy consistently, the more
166. 6:12opportunity you give the edge to
167. 6:13overcome short-term randomness. That's a
168. 6:16massive distinction, but we have to be
169. 6:18hyper definitive. The definition of the
170. 6:21terms that you use matters. Expectancy
171. 6:24is what actually matters. This is why
172. 6:26professional thinking isn't how often am
173. 6:29I right? It's what happens when I'm
174. 6:31right versus what happens when I'm
175. 6:33wrong. Imagine trader A wins 70% of the
176. 6:36time. That sounds amazing. He risks
177. 6:38$100, and when he wins, he makes $50.
178. 6:42When he loses, it's 100 bucks. Over 100
179. 6:45trades, 70 winners * $50 is $3,500.
180. 6:5030 losers * $100 is $3,000.
181. 6:54He only made $500 before trading costs.
182. 6:57Now, imagine trader B only wins 30% of
183. 7:00the time. And that sounds terrible. Same
184. 7:03$100 risk, but winners make him $300.
185. 7:0730 winners, $300, nine grand. 70 losers,
186. 7:11$100, $7,000.
187. 7:14Net result, he's up $2,000.
188. 7:17The trader who was wrong 70% of the time
189. 7:20made more money using this
190. 7:22oversimplified example. That's why
191. 7:24asking, "What's the win rate?" without
192. 7:26asking, "What's the average winner
193. 7:28versus average loser?" is almost
194. 7:30meaningless. But, there's a catch.
195. 7:32Before everyone runs over to TradingView
196. 7:34and changes their take profit to 10
197. 7:36times their stop loss, please don't do
198. 7:38that. That's not what I'm saying. You
199. 7:39can't manufacture an edge by arbitrarily
200. 7:42increasing your target because as you
201. 7:44move your target further away, your
202. 7:45probability of reaching it generally
203. 7:47goes down. The goal isn't maximum risk
204. 7:49to reward and it isn't maximum win rate.
205. 7:52It's finding a combination where the
206. 7:53probability of winning and the magnitude
207. 7:55of those wins produces positive
208. 7:57expectancy after cost of trading. And
209. 7:59then comes the hard part. Finding
210. 8:01something mathematically profitable is
211. 8:03one problem, but being psychologically
212. 8:05capable of trading it is another.
213. 8:08Personally, I think this is something
214. 8:09traders should consider much more
215. 8:11seriously. Two strategies can have
216. 8:13identical long-term profitability and
217. 8:15one might be completely impossible for
218. 8:17you to trade. One trader might happily
219. 8:18take tiny losses for weeks waiting for
220. 8:20that one monster trade. Another trader
221. 8:23could go insane after eight consecutive
222. 8:25losses. That person might perform much
223. 8:27better with higher win rate, lower
224. 8:29reward system. Neither of them is
225. 8:31automatically better than the other. The
226. 8:33best strategy isn't simply the one that
227. 8:35has the prettiest backtest. It's the one
228. 8:37with the edge that you, personally, can
229. 8:40actually continue executing when the
230. 8:42distribution becomes uncomfortable
231. 8:44because the market doesn't care that
232. 8:45you're frustrated. It doesn't know that
233. 8:47you lost 10 times in a row. And trade
234. 8:49number 11 doesn't become a winner simply
235. 8:51because you deserve one. And I think
236. 8:52this is one of the biggest lessons from
237. 8:54the data we've been analyzing over the
238. 8:55last couple of days. AI helps me find
239. 8:57patterns in hundreds of trades extremely
240. 8:59quickly, but the numbers expose
241. 9:01something much more interesting than
242. 9:02which day of the week performs best.
243. 9:04They show just how psychologically
244. 9:06difficult the mathematical variable
245. 9:08strategy can actually be to trade. I
246. 9:10don't think I could handle 47 losses in
247. 9:12a row. A strategy that can lose 88% of
248. 9:15its trades with disgusting losing
249. 9:17streaks. It would literally make you
250. 9:19feel like an idiot if you were to
251. 9:21continue trading it. How are you
252. 9:22supposed to keep that idea in the back
253. 9:23of your head knowing that this has a
254. 9:25positive expectancy? Meanwhile, there
255. 9:27could be another strategy that has a 70%
256. 9:30win rate and you feel like a genius
257. 9:31because of it, at least 70% of the time,
258. 9:34but it barely makes any money. So, stop
259. 9:36asking only, "What's the win rate. Start
260. 9:38asking what's my average winner, what's
261. 9:41my average loser, and what's my
262. 9:42expectancy. What's the historical
263. 9:44drawdown, and what's my biggest losing
264. 9:47streak, and can I deal with that
265. 9:48emotionally? This is why it's so
266. 9:50difficult for me to suggest a specific
267. 9:53strategy that's good for everyone,
268. 9:56because everyone has their own risk
269. 9:58tolerance. And I know a majority of
270. 10:00traders fall in the bucket of couldn't
271. 10:02take 47 back-to-back losses. I
272. 10:05referenced yesterday's video a lot, so
273. 10:06if you haven't seen it, you can click
274. 10:08right here. And if you enjoyed this
275. 10:09video, share it with your friends. Thank
276. 10:12you so much for watching, and we'll see
277. 10:13you in the next one.