Skip to main content

Evaluating Stock Pickers on FinTwit: Luck, Skill, and Honest Reporting

FinTwit hosts countless accounts claiming to identify undervalued stocks. Some accounts boast of picking stocks that doubled. Others claim returns far exceeding the market. Many sell newsletters or courses claiming to teach their picking methods. A few might even have real skill. Most are either lucky, overestimating their ability, or deliberately misleading.

The problem is simple: it's nearly impossible to distinguish luck from skill in stock picking, and performance reporting on FinTwit is almost entirely unverified and unaudited. Someone could flip a coin on whether to be long or short a stock, get lucky with a few calls, and then claim genius. You would have no way to know. This asymmetry—between real skill (which is rare) and performance reporting claims (which are abundant)—makes FinTwit stock picking particularly treacherous.

Professional investment managers at major institutions are constrained by audits, regulations, and reputational risk. When they claim a 20% return, regulators can audit those returns. When FinTwit accounts claim the same, there's no verification. This section teaches you to evaluate stock picking claims critically and avoid the traps of performance bias and survivorship bias.

Quick definition: Real stock-picking skill is the ability to consistently identify mispriced securities in a way that produces returns above market averages after costs, across multiple market cycles, with documented evidence.

Key takeaways​

  • Most FinTwit stock picks are luck disguised as skill — stock picking is a zero-sum game where most participants underperform the market even before costs
  • Performance reporting on FinTwit is unverified and systematically biased — successful picks are publicized, failed picks are deleted or forgotten
  • Survivorship bias is severe — unsuccessful stock pickers disappear, leaving only the lucky survivors claiming genius
  • Backtesting is unreliable — accounts can data-mine historical patterns until they find rules that would have worked in the past, then claim the patterns will work in the future
  • Real track records require audited documentation — if it's not in an SEC filing, it probably doesn't mean what the account claims
  • Beware the Dunning-Kruger effect — accounts most confident about picks are often those with the least experience and smallest sample sizes

The Skill Problem in Stock Picking​

Stock picking is a zero-sum game. For every person who beats the market, someone else underperforms it. The average active investor underperforms passive index investors (after fees and costs). This is documented fact, confirmed by decades of academic research.

The Vanguard Group, which manages trillions in assets, found that over a 20-year period ending in 2020, 85% of actively managed funds underperformed a simple total stock market index. That's not 50% underperforming (which would be random). That's 85%. The outperformers are barely statistically distinguishable from lucky random selection.

Yet FinTwit is full of accounts claiming to pick stocks better than average. Statistically, this is impossible. At most, maybe 15% of active investors beat the market. The probability that the people who boast about it online are exactly the lucky few is vanishingly small.

The more basic issue: how would you tell the difference between someone who is skilled and someone who is lucky?

Imagine a coin-flipping competition. A thousand people flip coins for 10 rounds. Roughly 10 of them will get heads nine times out of ten. Those 10 lucky people will naturally claim they have a skill for coin-flipping. They might even write a book: "My System for Winning at Coins." But it's luck, not skill.

The same dynamic applies to stock picking. If ten thousand retail traders pick stocks for ten years, maybe ten of them will beat the market by a large margin by luck alone. Those ten will have the highest confidence and the most compelling stories. They'll be the ones with large FinTwit followings. They'll sell courses and newsletters claiming they have the system figured out.

To distinguish skill from luck, you need:

  1. Large sample size — at least 50-100 picks, not 5
  2. Long time period — at least 5-10 years, ideally across different market regimes
  3. Comparison to benchmarks — is the outperformance better than you'd expect from randomness?
  4. Risk adjustment — did they take more risk (which explains higher returns) or genuinely pick better stocks?
  5. Independent verification — was someone auditing the results, or just the account reporting their own returns?

FinTwit stock picks almost never meet these criteria. Most accounts have small sample sizes, short histories, no benchmarking, and self-reported results.

Performance Reporting Bias on FinTwit​

Even if a stock picker is honest about their results, multiple biases distort the picture.

Survivorship bias is severe. Ten years ago, maybe one thousand accounts were posting stock picks on FinTwit. Most have been deleted. The people posting picks now include mostly:

  • People who have been lucky
  • People who gave up but didn't delete their account
  • New people with small sample sizes (not yet disproven)
  • People who delete accounts when they're wrong and start new accounts

You're seeing a survivor population. The graveyard of failed accounts is not visible. So if you scroll FinTwit, every active stock-picking account appears more successful than the average stock picker, because unsuccessful ones disappeared.

Selection bias compounds survivorship bias. An account that picked one stock that went up 300% will highlight that pick. They might have picked 20 others that flopped. But the algorithm will show you the big winner, not the average return. You see exceptional outcomes, not representative outcomes.

Anchoring bias distorts perception. If an account called "The Stock Whisperer" picks a stock at $50, gets lucky that it rises to $200, and then never picks anything good again, they'll still have "The Stock Whisperer" following them and credibility from the one big hit. That one big hit anchors everyone's memory. The 0 return of all the other picks fade.

Backtesting bias is severe among accounts that share "systems" or strategies. Someone can look at 20 years of historical stock data and find patterns that would have made money if you'd followed them perfectly. But historical patterns don't reliably predict future prices. The person found rules that fit past data through curve-fitting, not rules that actually predict stocks.

Deletions. The most insidious bias: accounts simply delete failed predictions. They might post 30 calls, 8 are right, 22 are wrong. But they delete the wrong ones. Now their timeline shows only wins. New followers see a 100% accuracy track record. There's no way for you to know unless you screenshot every prediction (most people don't).

Vague post-dictions. An account posts "This tech stock looks promising" vaguely. It goes down. They post "I called that caution months ago" (they didn't, or their exact words were vague). They reframe history to match outcomes.

Red Flags for Misleading Stock Picking Accounts​

Certain patterns strongly indicate that the account is either using biased reporting or lacks real skill.

Very high historical returns. "I'm up 500% this year" or "My average stock pick is up 150%." Either they're lucky (not skilled), they're reporting biased (showing only wins), they're taking huge risks (borrowing margin), or they're lying. Real consistent outperformance of this magnitude is so rare that you should assume one of the negative explanations rather than assume genius.

Recent account creation. An account that's been picking stocks for six months can't have much data to distinguish luck from skill. If they're already claiming "I'm one of the best pickers on FinTwit," they probably lack the perspective to know how much variance luck introduces.

All or nothing positions. Some accounts take 100% concentrated bets on individual stocks. That's not skill—that's gambling. Real portfolio managers diversify because they know the future is uncertain. Accounts that go all-in on one stock are just increasing variance, not demonstrating skill.

No risk disclosure. Quality accounts will say "I use margin" or "I'm concentrated in one sector" or "I do options." These carry higher risk. Accounts that post only returns without disclosing risk are giving a distorted picture.

Selling services. An account that makes money picking stocks doesn't need to sell courses teaching you to pick stocks. They have a profitable trading operation. Accounts that make money selling courses about stock picking are making money from teaching, not from picking. Beware their incentive. They profit if you buy the course, not if you profit from it.

Unable to explain their reasoning. When asked "why did you pick this stock?" they give vague answers: "It feels good" or "The chart looks bullish" or "My system flagged it." Real analysis explains the mechanism: valuation is attractive, competitive position is improving, catalysts exist. Vague reasoning reveals lack of deep thinking.

One very public loss. Some accounts will post a massive pick that goes spectacularly wrong, then claim it's "a learning opportunity" or "options blew up but the thesis is still right." Real losses aren't learning opportunities—they're losses. When an account doesn't seriously grapple with being wrong, they're probably not learning anything.

Only long picks. An account that only ever recommends buying stocks is in a bull market bias. They can't short (maybe for platform reasons) or they believe every stock goes up eventually (which reveals poor analysis). Real stock picking includes both shorts and longs.

What Real Stock Picking Track Records Look Like​

If you want to evaluate someone's stock picking legitimately, what should you look for?

Audited results. If the person manages investor money, their results are audited by regulators. SEC Form ADV filings show audited performance. If they claim high returns but have no audited track record, that's a red flag.

Long time period. At least 5-10 years of data. Shorter periods don't tell you whether someone has skill or luck.

Multiple market environments. Did they outperform in bull markets? Bear markets? Flat markets? Someone who outperforms in bull markets but underperforms in bear markets is not demonstrating skill—they're just in a bull market.

Documented picks with dates. Every recommendation is timestamped with entry price and exit price. Returns are calculated from actual entry/exit, not from current price. The full list is public, not a curated selection.

Risk-adjusted comparison to benchmarks. Returns are compared to appropriate benchmarks (not cherry-picked), and risk-adjusted (accounting for volatility and drawdowns).

Explanation of edge. They can explain why they think they have an edge. Not "I'm just good at it" but "I analyze working capital cycles better than the market" or "I have domain expertise in semiconductor supply chains." The explanation is testable.

Track record held up during losses. When the account has a period of underperformance, they explain why (bad sector positioning, timing miss) and what they learned. They don't disappear or delete posts.

Real stock pickers (the rare ones who genuinely outperform) typically have much of this documentation. FinTwit accounts almost never do.

Building Reasonable Skepticism​

The best approach to FinTwit stock picking is skepticism combined with understanding of the statistics.

Assume that most accounts are lucky, not skilled. Assume performance reporting is biased. Assume survival bias means you're seeing the lucky few. Given these assumptions, it's rational to be deeply skeptical of any stock pick claims.

That doesn't mean ignoring all FinTwit stock picks. Someone might mention an interesting company you hadn't considered. That's valuable—they gave you an idea to research further, not a recommendation to act on their judgment.

If you're going to follow stock-picking accounts at all, follow accounts that:

  • Have been picking for 5+ years
  • Share complete lists of picks (not cherry-picked wins)
  • Acknowledge mistakes and explain what they learned
  • Don't sell courses or premium newsletters
  • Use diversified positions (not concentrated bets)
  • Explain their reasoning in detail
  • Are willing to say "I don't know" or "this is uncertain"

Even then, treat their picks as ideas to research, not as recommendations to follow blindly.

Real-World Examples: Skill vs. Luck in Stock Picking​

Example 1: The Accidental 300-Bagger (Luck)

A FinTwit account posted in 2018: "Small-cap semiconductor company being undervalued. Buying 10,000 shares at $0.50."

The company was acquired in 2021 for $15. The account's $5,000 investment became $150,000. They marketed themselves as a genius stock picker. But here's the reality:

  • The company's acquisition was somewhat random—multiple interested buyers, one happened to win
  • The account picked dozens of other small-cap stocks that went to zero
  • They got lucky on timing (bought before the bull run in semiconductors)
  • They got lucky on catalyst (acquisition at premium valuation)
  • They had small sample size (one big hit doesn't prove skill)

This is survivorship and selection bias. The account showed the big winner, hid the losses.

Example 2: The Audited Track Record (Real Skill?)

Compare to a portfolio manager with 15 years of audited returns. Their track record shows:

  • Average annual return of 13% (vs. S&P 500 average of 10%)
  • Outperformance in 11 of 15 years
  • Slight underperformance in bear markets (smaller drawdowns)
  • Complete documentation of every position, entry, exit
  • Explanation that their edge comes from deep analysis of balance sheet quality and identifying companies with improving returns on invested capital

This manager might actually have skill. The track record is long, audited, diversified across many picks, and consistent. But even here, you can't be 100% certain—luck could explain the outperformance. But the track record is at least legitimate evidence.

Example 3: The Deleted Losses (Selection Bias)

A FinTwit account posted 50 stock picks over two years. You scroll their timeline and see:

  • $AAPL pick from 2023—stock up 50%, they're boosting about it
  • $NVDA pick from 2024—stock up 200%, they claimed genius status
  • A few other winners

But you notice: the ratio of current posts to picks seems off. Did they make only a few picks? Or did they make many but delete the losing ones?

Without a public archive, you can't verify. This is the critical weakness of FinTwit: deletions make performance biased.

Common Mistakes in Following Stock Pickers​

Many retail investors make systematic errors with stock-picking accounts.

They follow the accounts with highest returns. But highest returns usually means luck (or survivorship bias), not skill. They follow the accounts with most followers. But popularity is driven by entertainment and boldness, not accuracy. They assume that past picks that worked will keep working. But stock picking is uncertain. Historical results don't predict future success.

They put too much weight on stock-picking ideas and too little on fundamental research. A stock picked by someone famous should still go through your own analysis. If you can't explain why you're buying it, you shouldn't buy it just because someone online recommended it.

They assume the account has disclosed all picks. Many accounts delete losses or simply don't post about picks that went wrong. You're seeing a biased sample.

They confuse confidence with accuracy. The most confident pickers are often the least experienced. Actual experts typically express humility about uncertainty. Yet the confident pickers are the ones with large followings.

FAQ: Stock Picking on FinTwit​

How do I tell if someone is a real stock picker or just lucky?​

This is the hard question. The honest answer: it's very difficult. You need to track them for years across multiple market environments. If you can't do that, assume luck until proven otherwise.

Should I follow stock picks from FinTwit accounts?​

You can follow them as idea sources. But don't act on them without your own research. Use FinTwit picks as starting points for analysis, not as trade recommendations.

Do the best stock pickers share their picks on FinTwit?​

Rarely. The best stock pickers usually manage investor money and have legal restrictions on what they can say publicly. They might share general analysis but not specific picks. The people sharing detailed picks are often either people who don't manage significant money (so their results are unverified) or people making money selling courses/newsletters.

Can I beat the market by following FinTwit stock picks?​

Statistically, probably not. The average active investor underperforms the market by the cost of fees and trading. Even if you found a skilled picker (rare), by the time they publish picks, you're late to act. Professional investors act faster.

What's the difference between stock picking and investment research?​

Stock picking claims to beat the market. Investment research helps you understand companies. You can do solid investment research (learning about a company) without claiming to beat the market. One is useful; the other is boastful.

Should I pay for a stock picking newsletter or course?​

If someone can beat the market picking stocks, why would they sell courses for $99/month? They'd have billions managing money or be trading their own account. The fact that they're selling courses suggests their picks don't actually outperform. They make more money selling the course than picking stocks.

Summary​

Stock picking accounts on FinTwit are heavily biased toward showing success and hiding failure. Performance reporting is unaudited and subject to survivorship bias, selection bias, and intentional deletions. Most accounts claiming high returns are lucky, not skilled. Real stock-picking skill is statistically rare and requires long track records, audited documentation, consistent outperformance across market environments, and realistic risk disclosure. When evaluating stock picks, assume luck until proven otherwise by years of consistent results. Use FinTwit picks as idea sources for your own research, not as recommendations to follow blindly.

Next​

→ Options FinTwit warning