The S&P Short Where a 7.8 Point Stop Turned 32 Points Into 4R
SkyAnalyst AI journal entry: US500 Short on Sep 1, 2026 closed +4.06R on TP2. Full workspace view, decision log, and AI reasoning, unedited.
Before the trade, meet the system.
SkyAnalyst is not one AI trader. It is four specialist agents, each with its own data pipeline, each maintaining state between evaluations, and each required to agree before a position is sized. They don’t chat in prose. They write structured messages to a shared state object that each reads on every evaluation cycle. That’s what makes the system auditable, and it’s what this case study will show, step by step, on a specific setup the trend agent almost passed on.
Of the three shorts SkyAnalyst took on September 1, this one captured the least ground and paid the most. The S&P short travelled 31.7 points, from 7660.7 to 7629. The Dow short the same morning travelled 190 points and the Nasdaq short 238. Yet this trade returned +4.06R against the Dow's +1.63R, because the risk band on this entry was 7.8 points and the risk band on the Dow was 116.3. That is the whole lesson, and it is the least intuitive arithmetic in trading. R is reward measured against what was risked to get it. A system optimising for R does not chase big moves. It hunts for places where the invalidation level is close enough to the entry that a modest move becomes a large multiple. On September 1 the market offered exactly one of those, and it lasted about five minutes.
Broad-based selling, and one very specific level
The S&P opened September under broad selling pressure, gapping below the prior day's low of 7665.1 and trading roughly half a percent under Friday's close.
The macro case was the same one driving the Dow and Nasdaq shorts that morning. The 10-year had pushed to 4.77 percent, above the prior day's high, compressing the equity risk premium. Brent had surged above $92.50, adding cost-push inflation risk. DXY was firming at 99.64. Gold was cracking lower rather than catching a bid, which in a selloff signals liquidation rather than flight to safety. ISM Manufacturing printed 54.6 against a 55.2 forecast, a miss that did nothing to soften the rate narrative.
Breadth was emphatic and deteriorating. NYAD sat at -692 against a 5-day EMA of -516.6, and the five-day trajectory read -346, then -465, then -853, then -692. The prior session had seen an extreme -1186 low. This is not a tape where a 500-stock index sustains a rally.
Then the market did what oversold markets do. Price bounced about 23 points off the session low at 7629.4 into the New York open, and stalled at 7663.6, precisely at the confluence of VWAP and the prior day's low, before rolling over into a tight consolidation near 7652.
Sell the Rally into VWAP and Prior-Day-Low Rejection. The pattern name is long because the location is the whole trade. Price had to fail at one specific confluence for this entry to exist, and it held that failure for about five minutes.
The divergence that decided it
VIX rose while the index rallied, moving from 14.93 at the prior close to 15.36, up nearly three percent, at the same time the S&P was bouncing off its session low. Rising volatility during a rally is not confirmation of it.
It is a warning, because the market is pricing more risk into a move that is supposed to represent relief. It flags long setups as suspect and marks the rally as a fade candidate rather than the start of a recovery. Breadth agreed at -692 against a 5-day EMA of -516.6.
Why the location mattered more than the direction
The rally did not die at a round number or at a level chosen after the fact. It stalled at 7663.6, precisely where VWAP met the prior day's low, two references a great many participants watch at once. Price tested that confluence, failed, and rolled over into consolidation near 7652.
That failure supplies the stop location, at 7668.5, just above a level the market had demonstrably defended within the preceding few minutes. Entry at 7660.7 against it is a 7.8 point risk band, where the normal-regime band with VIX at 15.36 would have been 15 to 20 points.
The stop is the trade
So the first target at 7644, which would have paid roughly 1R on a conventional stop, paid 2.14R here, and the second target at 7629, effectively a retest of the session low, paid 4.06R. Nothing about the price forecast was unusually good. The entry was placed where being wrong would be cheap.
Professional traders: tight stops are not a virtue in themselves, and a 7.8 point stop placed at random in this tape is noise bait that gets hit within minutes. SkyAnalyst does not favor any single strategy or a fixed stop distance. It reads the tape first, finds where invalidation actually sits, and places risk against that level rather than against a preset number.
US500 (S&P 500): NY AM Session Analysis
Tue, Sep 01, 2026-11:33 AM ET
Market Environment Summary
The S&P 500 is under broad-based selling pressure today, gapping below yesterday's low (7665.1) and trading ~0.5% below Friday's close (7691.6). The macro backdrop is firmly bearish: the 10-year yield has pushed to 4.77% (above yesterday's high), compressing the equity risk premium further; Brent crude has surged above $92.50 (cost-push inflation risk); DXY is firming at 99.64; and gold is cracking lower, suggesting a general risk-off repricing rather than a flight to safety. ISM Manufacturing PMI printed at 54.6 (miss vs. 55.2 forecast), ISM Prices at 71.1 (in-line), and JOLTS at 7.27M (slight miss), the data mix is stagflationary: growth softening while prices stay sticky, reinforcing the Fed's hawkish stance. Breadth is deeply negative, mutual/ETF fund outflows persist, and NAS100 and US30 are both trading below their respective yesterday lows.
Price bounced ~23 points off the session low at 7629.4 into the NY open, but the rally stalled at 7663.6, precisely at the VWAP/prior-day-low confluence zone, and has since rolled over into a tight consolidation near 7652. The bounce exhausted below every meaningful overhead level. This is a textbook "sell the rally" structure in a bearish trend day.
Directional Bias: Bearish Volatility: Normal (VIX 15.36, expanding from 14.93 close, rising VIX supports shorts)
Step 1: Breadth & Volatility Regime
| Indicator | Reading | Interpretation |
|---|---|---|
| NYAD current | −692 | Deeply negative, well below zero |
| NYAD 5-day EMA | −516.6 | Deteriorating trend: −346 → −465 → −853 → −692 |
| NYAD yesterday | −853 (close), low −1186 | Yesterday saw extreme negative breadth |
| NYAD today range | −809 low / −220 high | Bounced off lows but still deeply negative |
| VIX | 15.36 (from 14.93 close) | Rising +0.43, now above 5d EMA of 15.05 |
Key findings:
- NYAD confirms bearish. The 5-day trajectory is deteriorating (−346 → −465 → −853), and today's reading at −692 sits well below zero. Breadth is not supporting any sustained rally in this 500-stock index.
- VIX is rising while SPX is bouncing: this is a classic divergence warning. VIX moved from 14.93 to 15.36 today (+2.9%) even as SPX bounced off the lows. This flags any long setup as suspect and confirms the short bias. VIX at 15.36 places us in the "normal" regime → 15-20pt stops appropriate.
- ⚠️ Critical flag: VIX rising + SPX bouncing = reversal warning for longs. Long confidence downgraded.
Step 2: Agent Synthesis
| Agent | Direction | Confidence | Key Detail |
|---|---|---|---|
| Macro Agent | BEARISH | 82% | Thin risk premium, fund outflows, stagflationary data |
| Trend Agent | BEARISH | 68% | Moderate, regime TRANSITIONING, reduce size |
| Agreement | ✅ Both bearish | , | But Trend Agent flags bounce risk, lower confidence |
- Macro Agent (bear 82%): Directly applicable, SPX is the risk barometer. Elevated yields (4.77%), hawkish Fed, fund outflows (SPY/IVV), and weak ISM data all reinforce selling rallies. ADP tomorrow 8:15 AM ET is next catalyst.
- Trend Agent (bear 68%, transitioning): Price remains below 60m fast EMA (7665), slow EMA (7681), VWAP (7663), and yesterday's low (7665). However, 5m/15m momentum has turned constructive off the 7629.4 low, so the regime is transitioning, hence reduced sizing is advised. Invalidation at 7683.
- Synthesis: Both agents bearish = directional agreement, but Trend Agent's lower confidence and TRANSITIONING regime means we should favor selling rallies into resistance rather than chasing new shorts at the lows. The 7663-7683 overhead supply zone is the key area to watch.
- Risk events: All high-impact data for today (ISM, JOLTS) released at 10 AM ET, already absorbed. No further high-impact USD events today. ADP tomorrow pre-market. Clear to trade.
Step 3: Gap & Daily Structure
| Level | Price | Significance |
|---|---|---|
| Prior day close | 7691.6 | Gap reference |
| Prior day high | 7713.2 | Remote resistance |
| Prior day low | 7665.1 | Key overhead resistance (broken, now resistance) |
| Daily pivot | 7689.97 | Overhead |
| 5-day EMA | 7694.5 | Overhead |
| Today's high | 7698.9 | Pre-market high, remote |
| Today's low | 7629.4 | Session support / target |
| 7650 round | 7650 | Congestion zone (current) |
| 7600 round | 7600 | Extended target |
Gap analysis: Current price 7652.7 vs. prior close 7691.6 = −38.9 pts / −0.51% gap down. This exceeds the 0.5% threshold on news (weak ISM + rising yields), suggesting the gap may continue rather than fill. The failed rally to 7663.6, which stopped well short of the prior close, confirms gap continuation bias.
Structure: Price broke below yesterday's low at 7665.1 during London, bounced to 7663.6 in the NY open (testing that broken support as resistance), and was rejected. Now consolidating 7644-7655. The prior day low at 7665.1 has flipped to resistance, this is the defining level for the session.
Step 4: Multi-Timeframe Technicals
60-Minute (Bias Frame)
- EMA alignment: Price (7652.8) < EMA fast (7665.2) < EMA slow (7681.5) → fully bearish cascade
- VWAP: 7663.5, price well below → bearish
- RSI: 39.1, neutral-weak, coming off oversold (26.8), no bullish divergence
- MACD: Line −12.89, below signal −10.59, histogram −2.3 (weakening bearish momentum but still firmly negative)
- Volume: High-volume spike on the 7629.4 low candle (7,248 ticks vs. 3,482 avg), climactic selling followed by relief bounce
- Verdict: Bearish structure intact. The bounce is a retracement within a downtrend, not a reversal. RSI exited oversold but hasn't generated any bullish divergence.
15-Minute (Confirmation Frame)
- EMA: Price above fast EMA (7649.7) but below slow EMA (7659.1) → mixed/transitional
- VWAP: 7664.0, price below → bearish
- RSI: 51.1, neutral, momentum exhausted from the bounce
- MACD: Line −1.74, above signal −4.49, histogram +2.74 → bullish crossover in progress (bounce momentum), but MACD line still below zero
- Fib retracement (7629.4 → 7655.1): Price sitting near the 23.6% level at 7649, the bounce has retraced minimally
- Verdict: The bounce has stalled. 15m shows improving momentum but unable to push above VWAP (7664) or prior day low (7665). Price is capped.
5-Minute (Precision Frame)
- EMA: Price (7652.8) above fast EMA (7650.2) and slow EMA (7648.5) → short-term bullish (bounce structure)
- VWAP: 7664.0, price ~11 pts below → bearish
- RSI: 55.2, neutral, fading from the 65.8 high at 14:40 UTC
- MACD: Line 2.07, below signal 2.48, histogram −0.40 → momentum rolling over, MACD about to cross bearish
- Key structure: NY session high at 7663.6, session low at 7631.5 (7629.4 London low). Price consolidating in the lower half of the NY range.
- Fibonacci (5m pivot 7644.1 → 7663.6): Currently at 55% retracement, mid-range, no directional edge from fib alone
- Verdict: The 5m bounce is losing steam. MACD histogram turning negative, RSI fading, volume declining. Price is compressing between 7644 support and 7655 resistance, setting up for a directional break.
Multi-Timeframe Summary
| Timeframe | Bias | Key observation |
|---|---|---|
| 60m | Bearish | Full EMA cascade, below VWAP, below all structure |
| 15m | Bearish-to-neutral | Below VWAP/slow EMA, bounce momentum fading |
| 5m | Neutral → turning bearish | MACD rolling over, capped at 7655, volume declining |
Step 5: Setup Filtering & Trade Proposal
Confluence Checklist: SHORT Setup (Sell Rally into 7663-7665 Zone)
| # | Confluence Factor | Status | Detail |
|---|---|---|---|
| (a) | Multi-TF EMA alignment | ✅ | 60m fully bearish; 15m below slow EMA; 5m losing upside momentum |
| (b) | Price below VWAP | ✅ | Below VWAP (~7664) on all timeframes |
| (c) | Prior day level interaction | ✅ | Yesterday's low 7665.1 now acting as resistance; already rejected once at 7663.6 |
| (d) | Both agents agree | ✅ | Macro bear 82% + Trend bear 68% |
| (e) | NYAD confirming | ✅ | −692, deteriorating 5-day trend |
| (f) | VIX aligned | ✅ | Rising (14.93 → 15.36), supports short bias |
Confluences: 6/6 ✅: Maximum confluence for a short setup.
Risk Assessment
- Trend Agent invalidation: 7683, our stop must not exceed this
- Trend Agent regime: TRANSITIONING, reduce size advised
- VIX regime: Normal (15.36) → 15-20pt stops appropriate
- Structural stop zone: Above VWAP (7664) + prior day low (7665.1) + buffer = 7668 provides clean structural invalidation. This is well within the Trend Agent invalidation of 7683.
- Risk: 7668 − 7663 entry = ~5pt stop if entering at ideal zone... but realistic entry zone is 7660-7664, making stop ~4-8pts from entry. Let me calculate precisely below.
Why Not Long?
Despite the 5m bounce structure, longs fail the filter:
- VIX rising while SPX bouncing = reversal warning (downgrade longs)
- NYAD deeply negative = breadth not supporting
- Price below VWAP on all timeframes
- 60m structure fully bearish
- Only 1-2 confluences for a long (5m EMA alignment, that's it)
- No long trade.
TRADE SETUP
Setup #1: US500 SHORT, Sell Rally into VWAP / Prior Day Low Rejection
-
Direction: SHORT
-
Narrative: Price gapped down 0.51% on stagflationary data (weak ISM + sticky prices + rising yields). The gap has continuation characteristics. The relief bounce from 7629.4 was rejected at the VWAP/prior-day-low confluence (7663-7665) and momentum is fading. This is a "sell the retest of broken support" setup, the most reliable pattern in a trend day.
-
Entry Zone: 7660-7665 (ideally on a retest of VWAP ~7664 or prior day low 7665.1)
-
Entry Trigger: Price tags 7660-7665 zone and prints a 5m bearish rejection candle (upper wick / engulfing / doji reversal) OR price breaks below the 5m ascending trendline support (~7648-7650 zone) with momentum, confirming the bounce has failed.
- Alternative (aggressive): If price fails to reach 7660+ and instead breaks below 7644 (today's 15:00 UTC low / 5m pivot low), enter short on the break with a stop above 7656.
-
Stop Loss: 7668.5 (3.5 pts above prior day low 7665.1, above VWAP, includes 1-2pt slippage buffer for automated execution). Well within Trend Agent invalidation of 7683.
-
Risk from ideal entry (7664): ~4.5 pts
-
Risk from conservative entry (7660): ~8.5 pts
-
Targets:
- TP1 = 7644: Today's NY afternoon low / 5m pivot support (structural). From 7664 entry = 20 pts = ~4.4R. From 7660 entry = 16 pts = ~1.9R. ✅
- TP2 = 7629: Today's session low (London low, the day's defining support). From 7664 = 35 pts = ~7.8R. From 7660 = 31 pts = ~3.6R. ✅
- TP3 = 7610: Extension target below session low if breakdown occurs (round number approach toward 7600). From 7664 = 54 pts = ~12R. From 7660 = 50 pts = ~5.9R. ✅
-
R-Multiples (from 7664 entry / 7668.5 stop):
- TP1 (7644) = 4.4R
- TP2 (7629) = 7.8R
- TP3 (7610) = 12.0R
-
R-Multiples (from aggressive 7644 break entry / 7656 stop):
- TP1 (7629) = 1.25R (at structural level ✅)
- TP2 (7610) = 2.8R
-
Suggested Management: Book 50% at TP1, trail stop to entry on remainder, book 30% at TP2, let 20% run toward TP3 with a trailing stop.
-
Position Sizing: Given TRANSITIONING regime and Trend Agent's "reduce size" guidance, size at 0.5-0.75% risk rather than the standard 1%. This is a high-confluence setup but the bounce dynamics warrant caution.
-
Quality Score: 8.5/10
-
Confidence: HIGH: 6/6 confluences met
Supporting factors:
- ✅ 60m EMA bearish cascade, 15m below VWAP/slow EMA, 5m momentum fading
- ✅ Price firmly below VWAP on all timeframes (~7664)
- ✅ Prior day low 7665.1 acting as resistance, already rejected once (7663.6)
- ✅ Both Macro (82%) and Trend (68%) agents bearish
- ✅ NYAD at −692, deteriorating 5-day trend
- ✅ VIX rising (14.93 → 15.36) aligned with short bias
- ✅ Gap >0.5% on news = continuation bias
- ✅ R:R exceptional (minimum 1.25R at worst entry, up to 12R at best)
Risks:
- ⚠️ Trend Agent regime = TRANSITIONING, 5m/15m showing bounce momentum, price may not reach the 7660-7665 entry zone and instead chop sideways
- ⚠️ 60m RSI was deeply oversold (26.8), oversold bounces can extend further before rolling; risk of a squeeze to 7670-7680 before resuming down
- ⚠️ ADP employment data tomorrow 8:15 AM ET, overnight positioning risk
- ⚠️ If NYAD improves sharply from −692 toward neutral, the breadth underpinning weakens
-
Trend Alignment: Bearish across 60m (primary), 15m (confirming), and fading on 5m. Daily structure fully broken (below yesterday's low, below 5d EMA). Macro confirms. This is a with-trend sell-the-rally setup, the highest probability pattern in a trend day.
Execution Notes for Automated System
- Primary trigger: Limit sell zone 7660-7665, activate on 5m bearish reversal candle close
- Alternative trigger: Sell stop below 7643.5 (1pt below 7644 support + 0.5pt buffer) if rally fails to materialize
- Stop: 7668.5 (hard stop, includes buffer)
- Time validity: Next 90 minutes (through ~1:00 PM ET). If not triggered, reassess, afternoon dynamics may shift.
- No-trade condition: If VIX drops below 15.0 and NYAD recovers above −300, the short thesis weakens, cancel pending orders.
Decision log
15:41 UTC, 58 percent, WAIT. The highest opening score of the three shorts taken that morning, which reflects how complete the bearish case already was: breadth deeply negative, VIX rising, yields above the prior day's high, the index below yesterday's low. What was missing was the bounce. Price was still near the session low, and selling there means a stop far enough away to ruin the arithmetic.
15:43 UTC, 52 percent, WAIT. Confidence fell as the relief rally developed, which is the correct direction for the number to move. A bounce in progress is the worst moment to short, and the score is measuring the entry rather than the thesis.
15:45 UTC, 52 percent, WAIT. Holding as price pushed into the VWAP and prior-day-low zone. The setup was arriving but the rejection had not printed, and without it there is no defensible place to put the stop.
15:46 UTC, 68 percent, ENTER. The rally stalled at 7663.6 and rolled over. Confidence jumped 16 points in one evaluation because the missing element had appeared: a tested and failed level. The Risk Agent placed the stop at 7668.5, 7.8 points above the 7660.7 entry, and sized accordingly. Five minutes of waiting converted a normal-regime 15 to 20 point stop into a 7.8 point one.
On a $100k account at 2.0% risk per trade.
Each trade risks +$2,000 (1R). The system's actual scale-out behavior may differ, see disclaimer.
| Scenario | R-multiple | Profit on $100k |
|---|---|---|
| Stop hit (invalidated) | -1R | −$2,000 |
| TP1 hitActual | +2.14R | +$4,280 |
| TP2 hit | +4.06R | +$8,120 |
| TP3 hit (max potential) (not tracked) | +0R | +$0 |
All six agents combined.
What this trade teaches
Compare the three shorts from September 1 side by side and the point makes itself.
The Dow captured 190 points for +1.63R. The Nasdaq captured 238 points for +2.97R. The S&P captured 31.7 points for +4.06R. Ranked by distance travelled the order is exactly reversed from the ranking by return. Anyone selecting trades by how far price moved would have ranked this one last.
The variable doing the work is the denominator. A 7.8 point stop is only available for a few minutes, at one level, after a specific event has confirmed it, and it is available precisely because the system waited through three declining confidence scores rather than selling the low.
The second lesson is that tight stops are not a virtue in themselves. A 7.8 point stop placed at random in this tape gets hit by noise almost immediately. What makes it defensible is that it sits above a confluence level the market tested and rejected in the preceding minutes. The tightness is a consequence of the location, never the goal.
From the desk
Three shorts, one macro read, three winners, and three completely different entry structures. The Dow cleared on a single evaluation because it offered a clean retest immediately. The Nasdaq needed nine because it was mid-collapse and had to be allowed to bounce first. The S&P needed four and produced the best R of the three off the smallest move.
That variety is the argument against a fixed playbook. The same bearish thesis on the same morning demanded a fade in one instrument, patience through an oversold bounce in another, and a five-minute window at a VWAP confluence in the third.
September opens after the system's first losing month since the January 12 inception. August closed at -2.67R across 44 trades, with the damage concentrated in a single book: GBPUSD long lost 6.18R over ten trades while the desk's short books collectively finished the month up 3.52R. On August 27 we split every instrument trader into independent long and short books so that a one-sided failure can no longer average itself into invisibility. The US500 short is now its own book, reported separately from US500 long, and September opened with it doing exactly what it is meant to do.
At a Glance
What this teaches about AI-driven trading
How is +4.06R possible on a 32 point move?
R measures reward against the risk taken rather than distance travelled. The stop sat 7.8 points from entry, so every 7.8 points of favourable movement equals 1R, and the 31.7 points from 7660.7 down to 7629 is 4.06 of those units. This is why a much larger move on a much wider stop can and often does pay less, as the Dow short the same morning demonstrates.
What does +4.06R mean in dollars?
R is the trade's risk unit, the distance between entry and stop. On a $100,000 account risking 2% per trade, 1R is $2,000, so +4.06R is roughly $8,120. Reporting in R rather than dollars keeps results comparable across account sizes and across instruments whose point values differ. The simulation panel in this article shows the same trade at several account sizes.
Is a 7.8 point stop on the S&P not far too tight?
It would be if it were placed arbitrarily, and a stop that tight dropped at a random price in this tape gets taken out by noise almost immediately. This one sits just above 7663.6, where the relief rally tested the VWAP and prior-day-low confluence and failed within the preceding few minutes. If price reclaimed that level the reason for the trade was gone, so a wider stop would only have made being wrong more expensive.
Why did confidence fall from 58 to 52 before rising to 68?
The score measures the quality of entering at that moment, not the strength of the thesis. It fell as the relief rally developed, because selling into an active bounce is a poor entry regardless of how bearish the backdrop is. It then jumped sixteen points in a single pass when the rally failed at resistance, because that one event supplied both the confirmation and the stop location that had been missing.
How do you count R on this trade?
This case study reports full potential, the R distance to the furthest target price reached, here the second target at 7629. Our weekly, monthly and year-to-date recaps use a stricter TP1 baseline that credits only the first target, so the same trade appears smaller in those totals. See the [August monthly recap](/blog/monthly-recap-2026-08) and the [2026 year-to-date review](/blog/ytd-2026).
Run your markets with SkyAnalyst
21-day free trial. No credit card. Full access to the Trend Agent, Macro Agent, and six-factor confluence scoring.
Trading involves substantial risk of loss. Past performance is not indicative of future results. The analysis shown was produced by an AI model operating on SkyAnalyst’s live trading infrastructure; it is shared for educational and research purposes only and is not financial advice. About reported results. Every AI Trader publishes three take-profit targets (TP1, TP2, TP3) per trade. The broker closes 100% of the position at TP1, so two distinct R-multiples appear in this article. The hero R-multiple is the full-potential R: where the market actually traveled (the highest take-profit hit, or the stop loss) before the setup was invalidated or exhausted. The realized R, shown on the TP1 row of the simulated returns panel, is TP1’s R (or -1R on a stop out). The realized R is what we log to our running track record. Both numbers are honest. Showing both is what lets readers see the full arc of the move and the conservative ledger entry it produced. Simulated returns in this article are calculated against a hypothetical $100,000 account at 2% risk per trade (1R = $2,000). These are educational reference figures and do not reflect any specific account or broker execution. Your actual result depends on your position size, your risk parameters, and live market conditions.
From the SkyAnalyst Journal
The Nasdaq Short That Said Wait Eight Times Before It Said Enter
Nine evaluations in twenty-seven minutes. Eight declined at 38 to 42 percent confidence while the macro case was already overwhelming. The ninth cleared at 68 percent, and price ran 238 points to target.
US30 Short, September 1: One Evaluation, One Entry, TP3
The Dow short that needed a single evaluation to clear. Breadth at -683, yields at 4.77%, and a retest that never gave back a tick. Entry to third target in 1 hour 17 minutes for +1.63R.
2026 So Far: +28.75R, and the Month That Changed How We Measure
Two hundred and twenty trades since the January 12 inception, 125 winners, +28.75R net. Seven green months and one red one, and the red one taught us more about the system than any of the seven that came before it.