From Ticker to Fair Value in 60 Seconds: Using EvidInvest's get_growth_trajectory MCP Tool

This post is for informational and educational purposes only. Nothing here constitutes financial advice. Always conduct your own research before making investment decisions.

What is the growth trajectory model?

Most DCF calculators ask you to pick a growth rate. That single number — pulled from a dropdown or typed in by hand — determines almost everything about the output. Change it by a few percentage points and the fair value swings by 30% or more. The problem is that a single static growth rate treats a company ripping through a product supercycle the same as one in structural decline. They are not the same.

EvidInvest's growth trajectory model solves this by classifying every stock's EPS growth pattern into one of five shapes before the DCF even runs. The classification is based on comparing near-term CAGR to longer-term historical averages across multiple time windows. The shape determines how long the model trusts the near-term growth rate before transitioning to the industry median.

The five shapes and their near-term durations:

ACCELERATING — near-term EPS growth is faster than the long-term average. The model treats this as a growth ramp in progress and uses g_near for 4 years.
STABLE — EPS growth is roughly steady across periods. No acceleration, no decay. The model uses g_near for 3 years.
RECOVERY — a 1-year spike while the 3-year CAGR is still depressed. The model treats recent momentum as short-lived and uses g_near for 2 years.
DECELERATING — past peak, with near-term growth slower than historical. The model shortens its trust window and uses g_near for 2 years.
DETERIORATING — multiple CAGR periods are negative. The model gives near-term growth almost no weight — g_near for 1 year only.

After the near-term window, the model transitions to g_long — the industry sector median — for the remaining years of the projection, then discounts everything to present value using the company's WACC. The result is a fair value range that adapts to where the company actually is in its growth cycle, not where you hope it is.

Prerequisites

You need two things before you can call get_growth_trajectory:

1. An EvidInvest API key. Sign up at evidinvest.com/developers. The free tier includes 100 API calls per day, which is enough for dozens of trajectory lookups and side-by-side comparisons.

2. The EvidInvest MCP server connected to your AI tool. For Claude Desktop, open settings, navigate to MCP Servers, and add the following configuration:

{
  "mcpServers": {
    "evidinvest": {
      "command": "npx",
      "args": ["-y", "@evidinvest/mcp-server@latest"],
      "env": {
        "EVIDINVEST_API_KEY": "your-api-key-here"
      }
    }
  }
}

Replace your-api-key-here with your actual key, save the config, and restart Claude Desktop. The same JSON block works in Cursor's MCP settings. You should see the EvidInvest tools appear in the tool list — get_growth_trajectory will be among them.

Step-by-step: calling get_growth_trajectory on NVDA

NVDA is the clearest current example of an ACCELERATING trajectory. Data center AI demand has pushed near-term EPS growth well above the semiconductor sector's long-run average, and the model recognizes that.

In Claude Desktop, type:

"What is NVDA's fair value and growth trajectory?"

Claude calls get_growth_trajectory("NVDA") and gets back a response like this:

{
  "symbol": "NVDA",
  "trajectory": {
    "shape": "ACCELERATING",
    "description": "Growth ramp in progress — near-term EPS growth faster than long-term average. Model uses g_near for 4 years before transitioning to industry rate.",
    "dur_years": 4,
    "g_near_pct": "45.0%",
    "g_long_pct": "15.0%",
    "growth_range_note": "Conservative assumes 15%/yr (industry median); bull assumes 45%/yr (near-term pace) sustained."
  },
  "fair_value_range": {
    "conservative": 83.50,
    "moderate": 281.20,
    "bull": 905.40,
    "mos_price": 140.60
  },
  "current": {
    "eps": 4.54,
    "price": 182.78,
    "valuation_status": "FAIR",
    "potential_return_pct": 53.6
  },
  "refreshed_at": "2026-05-10T08:42:33Z"
}

Here is what each field is telling you.

trajectory.shape: "ACCELERATING" — This is your DCF phase indicator. It tells you the model is trusting the near-term rate for a full 4-year window before reverting to the industry median. If this were STABLE, that window would be 3 years. If it were DETERIORATING, it would be 1 year. The shape changes how much of the valuation weight falls on today's elevated growth rate.

g_near_pct: "45.0%" — This is the weighted recent EPS CAGR. NVDA has compounded earnings at roughly 45% in its recent near-term window, driven by data center and AI chip demand. The model uses this rate for years 1–4 of the projection.

g_long_pct: "15.0%" — The semiconductor industry median long-run growth rate. After year 4, the model assumes NVDA's growth converges toward this benchmark. This is the conservative anchor that keeps the base-case valuation from getting untethered.

fair_value_range.moderate: 281.20 — The base-case intrinsic value. The model uses g_near for 4 years and g_long for the remaining terminal period, discounts everything at NVDA's WACC, and arrives at approximately $281 per share as the central estimate.

mos_price: 140.60 — The margin-of-safety entry price. This is fair_value_range.moderate × 50%. The 50% margin-of-safety is a classic value investing buffer that accounts for model uncertainty, analyst error, and macro risk. At $140 or below, you have a meaningful cushion relative to intrinsic value.

valuation_status: "FAIR" — NVDA at ~$182 sits between the mos_price of $140 and the moderate intrinsic value of $281. The model is not calling it a screaming buy — there is no 50% discount to intrinsic here — but it is also not overvalued relative to base-case assumptions.

potential_return_pct: 53.6 — The upside from current price to the moderate fair value estimate. At $182 with a base-case of $281, that is about 54% potential upside if the model assumptions hold.

The wide fair_value_range ($83 conservative, $905 bull) reflects the reality that ACCELERATING stocks carry the most valuation uncertainty. If the AI infrastructure buildout continues and NVDA sustains 45% EPS growth, the bull case is plausible. If growth normalizes faster than expected, the conservative case applies. The moderate estimate is the model's best single answer under base-case assumptions.

Contrast: calling get_growth_trajectory on INTC

Now call the same tool on INTC and watch what changes.

"What is INTC's fair value and growth trajectory?"

{
  "symbol": "INTC",
  "trajectory": {
    "shape": "DETERIORATING",
    "description": "Structural decline — multiple CAGR periods are negative. Model uses g_near for only 1 year.",
    "dur_years": 1,
    "g_near_pct": "-3.5%",
    "g_long_pct": "12.0%",
    "growth_range_note": "Conservative assumes 12%/yr (industry median); bull assumes -3.5%/yr near-term pace."
  },
  "fair_value_range": {
    "conservative": 12.40,
    "moderate": 31.80,
    "bull": 58.20,
    "mos_price": 15.90
  },
  "current": {
    "eps": -0.06,
    "price": 20.50,
    "valuation_status": "SPECULATIVE",
    "potential_return_pct": 55.0
  },
  "refreshed_at": "2026-05-10T08:42:33Z"
}

The DETERIORATING classification is grounded in Intel's actual reported numbers. FY2025 EPS came in at -$0.06. FY2024 EPS was -$4.38. Revenue contracted from roughly $63 billion to $53 billion over that same period. Multiple CAGR windows — 1-year, 3-year, 5-year — are all negative or deeply depressed. The model sees this pattern and responds by reducing the near-term trust window to a single year.

What that means in practice: the model applies g_near_pct of -3.5% for just 1 year, then assumes the business recovers toward the semiconductor industry median of 12%. The DCF is essentially saying: "We will not anchor much on what INTC has done recently, because multiple periods of deterioration make it unreliable. We will trust that it eventually converges toward sector norms."

The result is a fair value range of $12.40 to $58.20 — a spread of nearly $46 per share. That wide range is not a modeling flaw. It is an honest reflection of the uncertainty involved when a company's near-term trajectory is this unclear. The conservative case ($12.40) assumes the deterioration extends before recovery. The bull case ($58.20) assumes a faster turnaround. The moderate case ($31.80) is the base assumption that the industry median takes over relatively quickly.

valuation_status: "SPECULATIVE" — At $20.50, INTC is trading above the mos_price of $15.90 but well below the moderate fair value of $31.80. The SPECULATIVE label reflects that the upside is real ($31.80 - $20.50 = 55% potential return) but the path there requires a business recovery that has not yet materialized in the earnings data. The market is pricing in some probability of that recovery. The model is not assuming it.

What ACCELERATING vs DETERIORATING means for position sizing: When a stock is ACCELERATING and trading below intrinsic value — as NVDA was doing for much of 2024 — the model's confidence in near-term projections is higher and a full position allocation is more defensible. When a stock is DETERIORATING, the model is warning you that near-term earnings are unreliable guides. That is a signal to use a smaller speculative position, or to wait until the trajectory reclassifies as RECOVERY or STABLE before sizing up. The trajectory shape is not just a DCF input — it is also a risk sizing signal.

Advanced: side-by-side comparison in a single prompt

Once you understand what get_growth_trajectory returns, you can use it to make relative decisions, not just absolute ones. Try this prompt in Claude Desktop or Cursor:

"Compare NVDA and INTC fair values using get_growth_trajectory and tell me which offers better risk-adjusted upside."

Claude will call get_growth_trajectory twice — once for each ticker — and synthesize both responses into a single analysis. Because the tool returns structured JSON with consistent field names across all tickers, Claude can directly compare valuation_status, potential_return_pct, trajectory.shape, and fair_value_range without any additional calculations.

The output you get will reason over both datasets at once: NVDA's ACCELERATING shape and 54% upside to intrinsic vs. INTC's DETERIORATING shape and 55% upside to intrinsic — and explain that those two 55% figures are not equivalent. One is an upside to a model that trusts 4 years of recent growth. The other is an upside to a model that is mostly betting on mean reversion.

No manual Excel work. No rebuilding the DCF from scratch. No trying to hand-code a two-stage discount model in a spreadsheet. The structured JSON output from get_growth_trajectory is designed precisely for this kind of agent-driven reasoning — it gives Claude pre-computed inputs it can reason over, not raw data it has to model first.

EvidInvest's MCP server exposes 46 tools covering valuations, trajectory models, screeners, and portfolio analytics — see the full reference at evidinvest.com/developers.

Get started

If you have not connected EvidInvest MCP yet, this is the starting point:

Get your EvidInvest API key — free tier, 100 calls per day, no credit card required
See a live NVDA valuation in EvidInvest — the same data the get_growth_trajectory tool returns, rendered in the EvidInvest UI

The growth trajectory model is one of 46 tools in the EvidInvest MCP server. Once you have your API key and the server connected, the same pattern — call a tool, get structured JSON, let Claude reason over the result — applies across screeners, portfolio optimization, reverse DCF, peer comparisons, and more.

EvidInvest provides data and analysis tools for educational and informational purposes. This is not financial advice. Always conduct your own research before making investment decisions.