Limited buying: $2.6M bought (counted up to $5.0M); 7 distinct insiders. Holding it back: 365 days from first to last purchase.
12
Insider ScoreThe number from 0 to 100 shown on each cluster and used to rank the lists. By default it is the match score, plus the points of the AI review when there is one. It describes the buying; it is not a forecast or a recommendation.
<60 · limited buying
–
Since the cluster formed, vs benchmark
AI score not needed: the AI review runs from a match score of 40.
Score, not a recommendation.
Total bought$2.6M187 Form 4 purchases
Distinct insiders7open-market buyers
Insiders' average price$54.52no recent close
Cluster window2016-07-05 → 2017-07-05closed
partial dataAn input is missing, usually the market cap or the price history. A missing input counts as neutral in the score and is never estimated.Market cap is missing; missing inputs count as neutral in the score.
Price and insider purchases
Live chart from Yahoo Finance, loaded through this server; our daily closes when it is unavailable.
Points each input adds to the score, and how much of the points it could give it has
Input
Strength
Points
$2.6M bought (counted up to $5.0M)The dollars the insiders bought, a purchase reported jointly counted once, counted from $131. Each tenfold adds 12.4 points, up to $5.0M; more adds nothing. The bar is empty at $10.0K and full at $5.0M.
89%
+53
7 distinct insidersHow many different insiders bought. Each extra insider adds less than the one before (the points follow the log of the count: 4.3 per doubling). The bar is full at 10 insiders.
85%
+12
No CEO among the buyersWhether the chief executive is among the buyers: 10.6 points when yes.
0%
+0
Stock 6% below its 1-year highHow far the stock had fallen from its highest close of the past year, at the close before the buying. Every 10% of fall adds 1.9 points. A fall beyond 30% adds nothing more. The bar is full at a 30% fall.
19%
+1
365 days from first to last purchaseDays between the first and the last purchase. Each day takes off 0.11 points: buying close together keeps more. The bar is empty at 60 days.
0%
-41
Routine or plan purchases: 12% of the dollarsThe share of the dollars in routine or plan purchases (a Rule 10b5-1 plan, dividend reinvestment, a regular cadence). All of it routine would take off 5.6 points.
88%
-1
Largest buyer: LEMLEY FRASER (Director), 98% of the dollarsThe share of the dollars bought by the largest buyer. Buying spread evenly keeps the points; one buyer paying for everything takes off 12.7.
2%
-12
Match scoreA score from 0 to 100 worked out from the dollars bought, the number of insiders, whether a CEO bought, how far the stock had fallen, how spread out the purchases were and how much came from one buyer. Its weights were calibrated on past clusters. It is the default ranking of the lists and it decides which clusters get the AI review. It describes the buying; it is not a forecast.
12
Why are they buying?
AI score not needed: the AI review runs only on clusters with a match score of 40 or more. The filings in the table below are the primary evidence.
Transactions
126 Form 4 trades · 7 insidersDownload CSVUp to 5,000 rows
Purchases in this cluster, grouped by insider, largest first
Date
InsiderAn officer or director of a company, or an owner of more than 10% of its shares. The law requires them to report their trades in the company's stock to the SEC.
A buying runPurchases at one company with no pause longer than the cluster rule allows. A long pause ends the run, so one company can have several. ends after a pause in the buying, so one company can have several. Returns are the stock's return minus its benchmarkThe index a stock's return is compared with: a small-cap index for small companies, a large-cap index otherwise. (IWM for small companies, SPY otherwise), from the first open after the buying run became a cluster. Past results, not a forecast.
After this buying run became a cluster, vs benchmark:
21 days +5.0% (+3.3% raw)63 days +20.1% (+20.9% raw)126 days +33.4% (+42.9% raw)
Other insider buying runs at this company, newest first, with the return against the benchmark that followed