Weekly Selection | SECT Capital Quant Research

Redacted — proprietary method withheld. The space is kept so you can see where detail exists.

These are the 10 names I am backing for the week of September 21, 2026, and below each one I show my work: the momentum, the technicals, the sentiment, the fundamentals, and the macro I am leaning on. I do not ask you to take the picks on faith — I ask you to read the evidence and disagree if you can. The pillar weights and a few model internals stay redacted because they are the edge; everything that can be shown, is.

Macro environment

VIX

15.44

10Y−2Y

+0.25

HY OAS

2.70

Unemployment

4.1

SPY 20d

-0.91%

Here is how I read the backdrop: volatility is cheap at 15.4 — barely 0.97% a day — and cheap vol is exactly where I watch crowding hardest, the 10y−2y curve has steepened to +0.25, a quieter vote of confidence in duration, high-yield spreads at 2.70 are still contained, so credit is not yet contradicting the tape, and the S&P's trailing month is -0.9%, the trend I am actually trading against.

Volatility is the first thing I price, because it sets the size of every other bet. I scale gross exposure inversely to realised vol — target the risk, not the notional — so a calmer tape lets the book breathe and a violent one pulls it in automatically:

\sigma_{\text{daily}}\approx\dfrac{\mathrm{VIX}}{\sqrt{252}}\qquad w_{\text{gross}}\;\propto\;\dfrac{\sigma_{\text{target}}}{\sigma_{\text{daily}}}

Source: fred · as of 2026-09-20

How I make the call

I score every candidate from 0–100 on several independent pillars, then collapse them into one composite. The weighting is the part I keep back, but the structure is simple:

\text{Composite}(x) \;=\; \sum_i w_i \, S_i(x)

where \(S_i\) is a pillar score and \(w_i\) its weight. The exact pillar weights are mine to keep, so they are withheld; what I will tell you is that a neutral-pillar guard drops any pillar sitting at 50 (no signal) and renormalises the rest, so a silent factor never dilutes a loud one.

For the expected weekly return I do not trust a raw historical average at a given score — small samples lie. I shrink it toward a prior, blending the realised returns of past names that scored nearby with that prior in proportion to how much evidence I actually have:

\hat{r}(s) \;=\; \frac{n\,\bar r_{\text{near}} + k\,\pi(s)}{n + k}

Here \(k\) (the shrinkage strength) and the prior \(\pi\) take proprietary values , and they are exactly what I re-estimate as new weeks land — this is where the framework learns. The standard indicator formulas below I show in full, because none of that is the edge.

Indicator formula reference

RSI: $$\mathrm{RSI}_{14}=100-\frac{100}{1+RS},\quad RS=\frac{\overline{\text{gain}}_{14}}{\overline{\text{loss}}_{14}}$$

MACD: $$\mathrm{MACD}=\mathrm{EMA}_{12}-\mathrm{EMA}_{26},\quad \text{signal}=\mathrm{EMA}_9(\mathrm{MACD})$$

ATR: $$\mathrm{ATR}_{14}=\tfrac{1}{14}\sum TR,\quad TR=\max\big(H-L,\,|H-C_{-1}|,\,|L-C_{-1}|\big)$$

Bollinger: $$\mathrm{BB}=\mathrm{SMA}_{20}\pm 2\,\sigma_{20}$$

Cross-sectional z: $$z=\frac{x-\mu}{\sigma}$$

The picks

#1 SMTC SMTC Semiconductors Composite 65.2 / 100

Decision & pillar scores

Pillar	Score	Weight
Momentum	60.8
Catalyst	50.0	n/a
Sentiment	69.2
Technical	60.9
Fundamental	65.6
Smart Money		n/a
ML Ensemble
Composite	65.2	—
Universe percentile	99

Technicals

No indicators on file yet.

Sentiment & news

Fundamentals

P/E	51.4	PEG	n/a
P/B	13.66	P/S	16.44
ROE	51.2	ROA	21.9
Net margin	n/a	Rev growth	25.4
Debt/Equity	n/a	EPS growth	1063.4

SMTC selected based on: • UPCOMING CATALYSTS: High probability events within 30 days • COMPOSITE SCORE: 65.2/100

#2 NTSK NTSK Technology Composite 69.1 / 100

Decision & pillar scores

Pillar	Score	Weight
Momentum	60.6
Catalyst	50.0	n/a
Sentiment	58.6
Technical	65.3
Fundamental	59.6
Smart Money		n/a
ML Ensemble
Composite	69.1	—
Universe percentile	96

Technicals

No indicators on file yet.

Sentiment & news

Fundamentals

P/E	n/a	PEG	n/a
P/B	28.50	P/S	8.47
ROE	n/a	ROA	-53.0
Net margin	n/a	Rev growth	n/a
Debt/Equity	n/a	EPS growth	n/a

NTSK selected based on: • UPCOMING CATALYSTS: High probability events within 30 days • COMPOSITE SCORE: 69.1/100

#3 PURR PURR Biotechnology Composite 54.9 / 100

Decision & pillar scores

Pillar	Score	Weight
Momentum	54.1
Catalyst	50.0	n/a
Sentiment	62.0
Technical	80.1
Fundamental	37.0
Smart Money		n/a
ML Ensemble
Composite	54.9	—
Universe percentile	91

Technicals (2026-09-18)

RSI(14)	68.8	ADX(14)	42.6
MACD	1.025	Signal	1.011
MACD hist	0.014	ATR(14)	1.15
SMA 20	11.84	SMA 50	8.91
SMA 200	n/a	Vol ratio	1.25
Boll upper	13.46	Boll lower	10.22
CCI(20)	240.9	MFI(14)	50.7

Sentiment & news

Fundamentals

P/E	n/a	PEG	n/a
P/B	0.85	P/S	354.36
ROE	n/a	ROA	n/a
Net margin	n/a	Rev growth	n/a
Debt/Equity	n/a	EPS growth	n/a

PURR selected based on: • UPCOMING CATALYSTS: High probability events within 30 days • COMPOSITE SCORE: 54.9/100

#4 GRDN GRDN Health Care Composite 55.5 / 100

Decision & pillar scores

Pillar	Score	Weight
Momentum	58.7
Catalyst	50.0	n/a
Sentiment	50.0	n/a
Technical	67.9
Fundamental	73.8
Smart Money		n/a
ML Ensemble
Composite	55.5	—
Universe percentile	89

Technicals (2026-09-18)

RSI(14)	62.8	ADX(14)	30.8
MACD	1.815	Signal	1.232
MACD hist	0.583	ATR(14)	1.51
SMA 20	41.27	SMA 50	40.50
SMA 200	36.51	Vol ratio	1.00
Boll upper	48.55	Boll lower	33.99
CCI(20)	76.6	MFI(14)	74.2

Sentiment & news

No sentiment or news collected yet.

Fundamentals

P/E	56.2	PEG	n/a
P/B	9.26	P/S	1.91
ROE	30.8	ROA	15.8
Net margin	n/a	Rev growth	10.3
Debt/Equity	n/a	EPS growth	10.9

GRDN selected based on: • UPCOMING CATALYSTS: High probability events within 30 days • COMPOSITE SCORE: 55.5/100

#5 PS PS Financial Services Composite 59.0 / 100

Decision & pillar scores

Pillar	Score	Weight
Momentum	64.0
Catalyst	50.0	n/a
Sentiment	56.5
Technical	77.1
Fundamental	53.9
Smart Money		n/a
ML Ensemble
Composite	59.0	—
Universe percentile	86

Technicals (2026-09-18)

RSI(14)	69.7	ADX(14)	25.2
MACD	1.393	Signal	0.674
MACD hist	0.719	ATR(14)	2.76
SMA 20	40.56	SMA 50	38.06
SMA 200	n/a	Vol ratio	3.79
Boll upper	46.82	Boll lower	34.29
CCI(20)	242.3	MFI(14)	75.6

Sentiment & news

Fundamentals

P/E	79.8	PEG	n/a
P/B	9.66	P/S	26.14
ROE	5.6	ROA	4.2
Net margin	n/a	Rev growth	n/a
Debt/Equity	n/a	EPS growth	n/a

PS selected based on: • UPCOMING CATALYSTS: High probability events within 30 days • COMPOSITE SCORE: 59.0/100

#6 INSW INSW Energy Composite 73.0 / 100

Decision & pillar scores

Pillar	Score	Weight
Momentum	56.6
Catalyst	50.0	n/a
Sentiment	54.3
Technical	64.6
Fundamental	59.1
Smart Money		n/a
ML Ensemble
Composite	73.0	—
Universe percentile	81

Technicals

No indicators on file yet.

Sentiment & news

Fundamentals

P/E	17.8	PEG	n/a
P/B	1.19	P/S	6.53
ROE	37.0	ROA	27.5
Net margin	n/a	Rev growth	57.4
Debt/Equity	n/a	EPS growth	225.8

INSW selected based on: • UPCOMING CATALYSTS: High probability events within 30 days • COMPOSITE SCORE: 73.0/100

#7 BGC BGC Financial Services Composite 52.1 / 100

Decision & pillar scores

Pillar	Score	Weight
Momentum	41.0
Catalyst	60.5
Sentiment	82.7
Technical	64.3
Fundamental	66.4
Smart Money		n/a
ML Ensemble
Composite	52.1	—
Universe percentile	71

Technicals (2026-09-18)

RSI(14)	58.5	ADX(14)	11.3
MACD	0.208	Signal	0.228
MACD hist	-0.021	ATR(14)	0.35
SMA 20	12.06	SMA 50	11.53
SMA 200	10.16	Vol ratio	2.49
Boll upper	12.32	Boll lower	11.79
CCI(20)	65.6	MFI(14)	59.8

Sentiment & news

Fundamentals

P/E	n/a	PEG	n/a
P/B	n/a	P/S	n/a
ROE	0.1	ROA	0.0
Net margin	0.1	Rev growth	0.3
Debt/Equity	1.60	EPS growth	0.2

BGC selected based on: • UPCOMING CATALYSTS: High probability events within 30 days • COMPOSITE SCORE: 52.1/100

#8 TNK TNK Energy Composite 64.2 / 100

Decision & pillar scores

Pillar	Score	Weight
Momentum	57.1
Catalyst	50.0	n/a
Sentiment	57.2
Technical	73.2
Fundamental	41.1
Smart Money		n/a
ML Ensemble
Composite	64.2	—
Universe percentile	69

Technicals (2026-09-11)

RSI(14)	73.8	ADX(14)	24.5
MACD	4.459	Signal	3.795
MACD hist	0.664	ATR(14)	2.91
SMA 20	90.98	SMA 50	81.25
SMA 200	71.54	Vol ratio	2.01
Boll upper	98.91	Boll lower	83.05
CCI(20)	224.2	MFI(14)	67.6

Sentiment & news

Fundamentals

P/E	n/a	PEG	n/a
P/B	n/a	P/S	n/a
ROE	0.2	ROA	0.2
Net margin	0.4	Rev growth	-0.2
Debt/Equity	n/a	EPS growth	-0.1

TNK selected based on: • UPCOMING CATALYSTS: High probability events within 30 days • COMPOSITE SCORE: 64.2/100

#9 RBLX RBLX Media Composite 38.1 / 100

Decision & pillar scores

Pillar	Score	Weight
Momentum	57.8
Catalyst	50.0	n/a
Sentiment	68.6
Technical	59.1
Fundamental	35.4
Smart Money		n/a
ML Ensemble
Composite	38.1	—
Universe percentile	56

Technicals (2026-09-18)

RSI(14)	61.7	ADX(14)	27.3
MACD	1.879	Signal	1.062
MACD hist	0.816	ATR(14)	2.25
SMA 20	43.12	SMA 50	43.89
SMA 200	57.81	Vol ratio	1.02
Boll upper	51.37	Boll lower	34.87
CCI(20)	81.4	MFI(14)	64.4

Sentiment & news

Fundamentals

P/E	n/a	PEG	n/a
P/B	144.18	P/S	6.99
ROE	-288.8	ROA	-10.6
Net margin	n/a	Rev growth	41.3
Debt/Equity	n/a	EPS growth	n/a

RBLX selected based on: • UPCOMING CATALYSTS: High probability events within 30 days • COMPOSITE SCORE: 38.1/100

#10 SPCX SPCX Telecommunication Composite 52.9 / 100

Decision & pillar scores

Pillar	Score	Weight
Momentum	45.0
Catalyst	50.0	n/a
Sentiment	55.7
Technical	71.5
Fundamental	45.7
Smart Money		n/a
ML Ensemble
Composite	52.9	—
Universe percentile	34

Technicals (2026-09-18)

RSI(14)	58.1	ADX(14)	24.1
MACD	4.146	Signal	3.727
MACD hist	0.420	ATR(14)	7.41
SMA 20	145.33	SMA 50	134.64
SMA 200	n/a	Vol ratio	2.53
Boll upper	156.82	Boll lower	133.84
CCI(20)	105.0	MFI(14)	61.1

Sentiment & news

Fundamentals

P/E	n/a	PEG	n/a
P/B	17.67	P/S	110.98
ROE	n/a	ROA	n/a
Net margin	n/a	Rev growth	n/a
Debt/Equity	n/a	EPS growth	n/a

SPCX selected based on: • UPCOMING CATALYSTS: High probability events within 30 days • COMPOSITE SCORE: 52.9/100

A standing note on method. I run this book in paper-trading mode, so every fill you see is simulated rather than a realised, audited track record — I would rather state that plainly than flatter the numbers. Nothing here is investment advice, an offer, or a solicitation; it is my own research, published so it can be read and argued with in the open. The blacked-out passages mark the parts of the process I keep proprietary. And because the framework recalibrates every week, where my read was wrong I expect the priors — not my ego — to be the first to say so.