Research / Track record

Sigma Estimates track record

Every estimate is locked before the company reports and scored against the reported figures afterwards. Nothing is edited or removed. The estimate that counts is the last one locked before the report day.

LIVE RECORD · SINCE NOV 8, 2026
Revenue · settled quarters
2
all comparable
Revenue · average error, Sigma
1.84%
median 1.84%
Revenue · average error, consensus
3.51%
median 3.51%
Revenue · Sigma closer
2 of 2
50% inside the Sigma range (target ≈ 50%)
EPS · settled quarters
2
all comparable
EPS · average error, Sigma
2.32%
median 2.32%
EPS · average error, consensus
5.12%
median 5.12%
EPS · Sigma closer
2 of 2
100% inside the Sigma range (target ≈ 50%)

Locked, waiting for the report

CompanyModelReportSigma revenueConsensusvs. consensusRangeSigma EPSLocked
PANWv2.1.1Nov 18, 2026$3.34B$3.31B+1.0%$3.32B–$3.38B$1.03Nov 12, 2026
DELLv2.1.1Nov 24, 2026$51.08B$49.25B+3.7%$50.57B–$52.13B$7.17Nov 17, 2026
WDAYv2.1.1Nov 24, 2026$2.71B$2.69B+0.9%$2.70B–$2.74B$2.95Nov 17, 2026
CRWDv2.1.1Dec 1, 2026$1.55B$1.53B+1.6%$1.54B–$1.57B$0.33Nov 17, 2026
MDBv2.1.1Dec 1, 2026$791.4M$762.5M+3.8%$787.4M–$800.0M$1.84Nov 18, 2026
MRVLv2.1.1Dec 1, 2026$3.19B$3.15B+1.0%$3.15B–$3.25B$1.11Nov 17, 2026
CRMv2.1.1Dec 2, 2026$11.53B$11.46B+0.5%$11.47B–$11.65B$4.10Nov 12, 2026
SNOWv2.1.1Dec 2, 2026$1.71B$1.65B+3.3%$1.70B–$1.73B$0.70Nov 13, 2026
ADBEv2.1.1Dec 9, 2026$6.92B$6.83B+1.3%$6.88B–$6.99B$6.47Nov 16, 2026
AVGOv2.1.1Dec 9, 2026$35.30B$34.90B+1.2%$34.94B–$36.03B$3.95Nov 17, 2026
MUv2.1.1Dec 23, 2026$64.43B$61.72B+4.4%$63.78B–$65.75B$43.87Nov 17, 2026
LRCXv2.1.1Jan 27, 2027$8.78B$8.63B+1.7%$8.69B–$8.96B$2.50Nov 17, 2026
METAv2.1.1Jan 27, 2027$76.51B$75.04B+2.0%$75.52B–$78.10B$8.90Nov 17, 2026
NOWv2.1.1Jan 27, 2027$4.41B$4.37B+1.0%$4.39B–$4.46B$1.11Nov 17, 2026
STXv2.1.1Jan 27, 2027$4.96B$4.82B+2.7%$4.91B–$5.06B$10.01Nov 17, 2026
AMDv2.1.1Feb 2, 2027$17.60B$17.08B+3.0%$17.42B–$17.96B$3.00Nov 17, 2026
QCOMv2.1.1Feb 3, 2027$11.04B$10.82B+2.0%$10.93B–$11.26B$2.31Nov 17, 2026
UBERv2.1.1Feb 3, 2027$15.71B$15.65B+0.4%$15.51B–$16.04B$0.96Nov 17, 2026
AMZNv2.1.1Feb 4, 2027$248.8B$245.2B+1.5%$245.6B–$254.0B$2.88Nov 17, 2026
KLACv2.1.1Feb 4, 2027$4.80B$4.72B+1.6%$4.75B–$4.89B$1.47Nov 17, 2026
RDDTv2.1.1Feb 4, 2027$1.09B$1.02B+6.6%$1.08B–$1.11B$2.25Nov 17, 2026
WDCv2.1.1Feb 4, 2027$5.26B$5.13B+2.6%$5.21B–$5.37B$6.50Nov 17, 2026
DDOGv2.1.1Feb 9, 2027$1.24B$1.20B+3.0%$1.23B–$1.25B$0.77Nov 17, 2026
APPv2.1.1Feb 10, 2027$2.46B$2.38B+3.2%$2.43B–$2.51B$4.85Nov 17, 2026
VRTv2.1.1Feb 10, 2027$4.62B$4.48B+3.1%$4.54B–$4.81B$2.59Nov 17, 2026
ABNBv2.1.1Feb 11, 2027$3.33B$3.26B+2.2%$3.29B–$3.40B$0.89Nov 17, 2026
AMATv2.1.1Feb 11, 2027$11.26B$11.09B+1.5%$11.15B–$11.49B$4.52Nov 17, 2026
ANETv2.1.1Feb 11, 2027$4.11B$3.97B+3.5%$4.07B–$4.20B$1.41Nov 17, 2026
TOSTv2.1.1Feb 11, 2027$1.99B$1.96B+1.2%$1.96B–$2.03B$0.36Nov 17, 2026
NVDAv2.1.1Feb 24, 2027$129.0B$125.3B+2.9%$127.6B–$131.6B$2.91Nov 18, 2026

Settled

CompanyModelReportedRevenueSigmaConsensusSigma errorConsensus errorCloserIn range
NVDAv2.1.1Nov 17, 2026$115.8B$112.1B$108.9B3.23%5.97%Sigmano
AMATv2.1.1Nov 12, 2026$10.41B$10.46B$10.30B0.45%1.05%Sigmayes

By model version

ModelFirst lockRevenue settledSigma errorConsensus errorSigma closerEPS settledEPS Sigma / consensus error
v2.1.1Nov 11, 202621.84%3.51%2 of 222.32% / 5.12%

A new model version only applies to estimates locked after it went live; earlier locks keep the version they were made with. v2.1 weights recent quarters more and states each estimate’s position against the company’s own guidance range.

By peer group (revenue)

Peer groupSettledSigma errorConsensus errorSigma closer
semiconductors & hardware21.84%3.51%2 of 2
BACKTEST · HYPOTHETICAL

Before going live each model version was tested on past quarters: statistical-v2.1, 30 companies and 359 quarters, average revenue error 2.39% for Sigma against 3.27% for consensus; statistical-v2.0.1, 30 companies and 360 quarters, average revenue error 2.40% for Sigma against 3.28% for consensus. Backtest results are hypothetical and do not show how the model performs live; the live record above does.

How this is scored. The Sigma estimate (statistical-v2, experimental; the version of each estimate is shown) starts from consensus and corrects it for each company’s record against consensus, pulled toward its peer group. Error is the absolute difference from the reported figure, as a percentage of it. Quarters where consensus and the report are not comparable (for example after a spin-off) are shown but excluded from the averages. The range is the 25th–75th percentile of past errors, so about half of outcomes should fall inside it. General research, not personalised investment advice. Methodology
© SigmaThesis · Not investment adviceAs of Nov 18, 2026 · 30 companies covered