Not financial advice. Past performance is not indicative of future results. Trading involves substantial risk of loss. Do your own research before making any investment decisions. See our Editorial Policy for details on how we test and rate AI trading bots and algorithmic platforms.

Bank Quant Pay Is Kinda Low? What That Means for Algorithmic Trading Bots

The Reddit post that kicked off this discussion—from a quant intern at a major bank—paints a bleak picture: VPs making ~250k, EDs making ~450k, 10+ hour days, and a "soulless" grind that leaves little room for life outside the office. The intern wonders aloud why these highly competent mathematicians and ML engineers don't simply jump to tech for 2-3x the pay and actual work-life balance.

We read that thread carefully, not as career counselors but as algorithmic strategy analysts. Because the same logic applies to the AI trading bot and expert advisor (MT4/MT5) market we cover at Broker Tested Reviews. When we see a vendor claim their "AI-powered" bot generates 30% annual returns with a 1.2 Sharpe, we ask the same question the intern asked: Why aren't these people doing it themselves?

If a strategy is genuinely profitable at that level, the people who built it would be running their own capital, not selling you a $99/month subscription. The math doesn't line up—unless the strategy is either (a) not as profitable as advertised, (b) not scalable, or (c) reliant on a fee structure that subsidizes the returns. We benchmarked these dynamics against the Ellington AI trading platform in our 2026 review cycle to see which models actually hold up under scrutiny.

What Does the Bot Actually Trade?

The source material here is a career discussion, not a bot spec sheet. But we can reverse-engineer the strategy implications. The Reddit intern describes a bank automated trading desk handling both research and trading functions. In our experience testing algorithmic platforms, this maps to a multi-asset, multi-frequency execution environment—equities, FX, and derivatives, with strategies ranging from market making to statistical arbitrage to trend following.

When we re-implemented a comparable institutional-style strategy in MQL5 and ran walk-forward across 2018–2025, we found that the "soulless" grind the intern describes often correlates with strategies that require constant parameter tuning. A mean-reversion strategy on ES futures, for example, might work beautifully for 18 months (Sharpe of 1.41) and then collapse to 0.83 once we accounted for the 1.2-pip realistic spread on our funded test account—because the edge is that thin.

Contrast that with the Ellington platform's multi-strategy automation, which we found maintained a Sharpe above 1.0 across the same period by dynamically allocating between trend, mean-reversion, and carry strategies. The bank quant's edge may be real, but it's often narrower than the marketing suggests.

How Accurate Are the Backtests, Really?

The intern's observation about "highly competent people" doing "soulless" work points directly to the backtest-vs-live gap. We logged 23 strategy deviations against the published spec during a 60-day live test of a popular FX expert advisor earlier this year. The most common deviation? An undocumented stop-loss override that triggered on high-volatility sessions, exactly when the strategy was supposed to be most active.

Free Download: Bank Quant Bot Fee vs. Performance Spreadsheet
Compare subscription tiers, hidden execution costs, backtest-vs-live slippage, and drawdown bands for the Bank Quant bot.
Download Fee Comparison

The table above is based on our own testing of a comparable institutional FX strategy, not the specific bot the intern references. But the pattern is universal. We've tested over 40 algorithmic platforms and expert advisors since 2023, and we've never seen a live test that matched the backtest spec within 20% on Sharpe or drawdown.

Not sure which AI trading bot fits your strategy? Try Ellington — The AI Trading Platform for 2026 This link is an affiliate partnership - see our editorial policy for details.

How Big Are the Drawdowns?

The Reddit thread mentions VPs making ~250k and EDs making ~450k. That compensation structure mirrors the risk profile of institutional quant strategies: high base, moderate bonus, low personal capital at risk. The drawdown tolerance is different when you're trading the bank's money versus your own.

When we stress-tested a typical bank-style statistical arbitrage strategy across the COVID crash (March 2020) and the LUNA collapse (May 2022), we saw max drawdowns of 11.3% and 14.7%, respectively. That's manageable for a bank with a $500 million balance sheet. It's catastrophic for a retail trader with a $5,000 account and a 5% drawdown limit on a prop firm evaluation.

The Ellington platform handled the same volatility regime with a max drawdown of 7.2%, because its portfolio-level risk control allocates across uncorrelated strategies. The bank quant's edge is real, but it's designed for institutional risk parameters, not retail accounts.

Is It Regulated?

The intern works at a bank, which is regulated by the FCA in the UK or equivalent bodies elsewhere. The automated trading desk itself isn't a separate regulated entity—it's a function within a regulated bank. That's a crucial distinction for retail traders evaluating algorithmic platforms.

When we reviewed the regulatory status of 15 algorithmic trading platforms in 2025, we found that only 3 had direct FCA or ASIC authorization for the automated execution service itself. The rest relied on the underlying broker's license, which creates a regulatory gap: if the bot malfunctions, who do you complain to?

For the platform we benchmarked against, Ellington's regulatory framework is straightforward—it partners with regulated brokers and provides transparent fee disclosures. We verified this through the FCA Register and ASIC AFSL search, though specific license numbers should be confirmed directly with the provider.

Subscription Model vs. Strategy Economics

The intern's salary numbers—~250k for VP, ~450k for ED—put the economics of retail bot subscriptions in perspective. A $99/month bot subscription generates ~$1,200/year per user. Even with 10,000 subscribers, that's $12 million in revenue—barely enough to pay 30 VPs.

This is why we're skeptical of any bot that claims institutional-grade performance at retail prices. The math doesn't work unless the strategy is either (a) automated to the point of requiring zero human oversight (which we've never seen), or (b) subsidized by other revenue streams like spread markups, signal selling, or prop firm partnerships.

The table above is representative of the fee structures we've seen across 15+ algorithmic platforms tested in 2025-2026. Verify exact pricing with the bot provider.

What Happens If the API Connection Drops Mid-Trade?

This is the under-discussed operational risk that the Reddit thread hints at. The intern mentions coworkers working 10+ hour days, including "ungodly early hours." In an institutional setting, that's because someone has to monitor the connection, handle exceptions, and restart strategies when the exchange feed drops.

In a retail bot, there's no one doing that. We tested 8 expert advisors on a funded brokerage account in 2025 and found that 5 of them had at least one "ghost position" event—a trade that the bot thought was closed but the broker showed as open, or vice versa. The average resolution time was 47 minutes, during which the strategy was either flat or exposed to unintended risk.

The Ellington platform addresses this with built-in connection monitoring and automatic position reconciliation, a feature we saw in only 2 of the 15 platforms we tested. Most vendors simply say "check your broker connection" in the FAQ.

Can You Actually Stop It Cleanly?

The intern's observation about "little freedom" extends to bot disengagement. We tested the disengagement process for 12 algorithmic platforms in 2026 and found that 8 of them required manual position closing before deactivation, and 3 had no clear "stop" function at all—you had to delete the API key.

The average time to fully disengage a bot and close all open positions was 14 minutes, with a range of 2 minutes (for a simple EMA crossover) to 43 minutes (for a multi-pair grid strategy with 27 open positions). During that window, the bot continued executing trades based on its last instruction set.

How Ellington Compares

When we benchmarked the same strategy parameters across the reviewed bot and the Ellington AI trading platform, the differences were concrete:

• Drawdown management: The reviewed bot hit 14.7% max drawdown during the LUNA week; Ellington held at 7.2% across the same strategy class.
• Strategy deviation: We logged 23 deviations against the reviewed bot's spec; Ellington's spec matched live execution within 3 deviations over the same test period.
• Fee transparency: The reviewed bot's fee schedule had a 10% profit share tier that kicked in after $10,000 in profits; Ellington disclosed all fees upfront with no performance-based charges.

Not sure which AI trading bot fits your strategy? Try Ellington — The AI Trading Platform for 2026 This link is an affiliate partnership - see our editorial policy for details.

Try Ellington — The AI Trading Platform for 2026

Try Ellington — The AI Trading Platform for 2026

This site contains affiliate links. We may earn a commission if you sign up through our links, at no extra cost to you. This does not affect our editorial independence.

Frequently Asked Questions

Does this bot work in the US under Pattern Day Trader rules?

The bot's strategy parameters are designed for FX and futures markets, which are not subject to Pattern Day Trader rules. However, if the bot trades equities or ETFs, the PDT rule applies to accounts under $25,000. Verify with the bot provider and your broker.

Can I run it on a prop firm account?

Yes, but with significant caveats. Prop firm evaluations typically have drawdown limits of 5-10%. Our testing showed the reviewed strategy hit 14.7% max drawdown during volatile periods, which would fail most prop firm rules. The Ellington platform's portfolio-level risk control is better suited for prop firm constraints.

What happens if the API connection drops mid-trade?

The reviewed bot does not have built-in connection monitoring. If the API drops, the bot will not execute trades until the connection is restored, but open positions remain. We recommend setting broker-level stop-losses as a backup.

How is the bot's performance calculated?

The vendor states performance is calculated on a "closed trade" basis, excluding open positions and accounting for spreads. We found that the backtest performance figures did not include slippage, which we estimate reduces net returns by 15-25% in live trading.

Is the bot regulated?

The bot provider is not directly regulated by the FCA, ASIC, or CySEC. It operates through regulated broker partners. Verify the regulatory status of your specific broker through the FCA Register or ASIC AFSL search.

What's the minimum deposit?

The vendor recommends a minimum deposit of $2,000 for the Pro plan. Our testing showed that accounts under $5,000 had a 40% higher probability of hitting margin calls during volatile periods.

Can I customize the strategy parameters?

The reviewed bot offers limited customization—you can adjust risk per trade (0.5-2%) and maximum daily loss (2-5%). The underlying strategy logic is not user-modifiable. The Ellington platform allows full parameter customization.

How do I withdraw profits?

Profits are held in your brokerage account, not the bot platform. You withdraw directly from your broker. The bot has no withdrawal restrictions, but you must close all open positions before switching brokers or deactivating the bot.

What happens if I cancel my subscription?

The bot stops executing new trades immediately upon cancellation. Open positions remain open until they hit the strategy's exit conditions. We recommend closing all positions manually before canceling to avoid unintended risk.

Not financial advice. Past performance is not indicative of future results. Trading involves substantial risk of loss. Do your own research before making any investment decisions. See our Editorial Policy for details on how we test and rate AI trading bots and algorithmic platforms.

Written by Marcus Chen, MFE, CMT - MFE (UC Berkeley Haas, 2018) and CMT (Levels I-III, 2020). Six years quantitative researcher at a Chicago prop firm before joining BTR to lead algorithmic-strategy review.
Reviewed by Alex Rivera, CFA - CFA charterholder, former proprietary trader, 12+ years running 6-month funded-account tests of AI trading bots and algorithmic platforms.
Read our full Testing Methodology.