```html
<div class="seastead-speed-estimate">
  <h3>Estimated steady-speed with 2× Shinmaywa 2.5&nbsp;m mixers (very rough)</h3>

  <p>
    <strong>What this estimate assumes (important):</strong>
  </p>
  <ul>
    <li><strong>Steady-speed condition:</strong> total thrust = total hydrodynamic drag.</li>
    <li><strong>Drag model:</strong> drag grows with speed squared: D = k·V².</li>
    <li><strong>Calibration point for drag:</strong> assumes your seastead’s “full power” top speed is about
      <strong>1.5 mph</strong> with both thrusters, using the table’s point (740 N per thruster at 1.5 mph).
      This sets the drag constant <em>k</em> for the rest of the calculations.</li>
    <li><strong>Thrust vs speed curve:</strong> uses your provided thrust table (linearly interpolated between points).</li>
    <li><strong>Power scaling:</strong> assumes thrust at a given speed scales approximately linearly with the commanded
      electrical power relative to 3.2 kW (i.e., 2.2 kW produces ~2.2/3.2 of the thrust curve, etc.).</li>
  </ul>

  <p>
    Because this is calibrated to the 1.5 mph / 740 N-per-thruster point, the results below are best interpreted as
    “relative performance vs power” <em>given that calibration</em>. Real-world speed can be significantly different if
    actual drag is higher/lower than this implied drag (columns, braces, cables, wave drag, added marine growth, etc.).
  </p>

  <table border="1" cellpadding="6" cellspacing="0">
    <thead>
      <tr>
        <th>Power per thruster (kW)</th>
        <th>Total electric power (kW)</th>
        <th>Estimated steady speed (mph)</th>
        <th>Estimated steady speed (kn)</th>
        <th>Estimated steady speed (m/s)</th>
        <th>Thrust per thruster at that speed (N)</th>
        <th>Total thrust at that speed (N)</th>
      </tr>
    </thead>
    <tbody>
      <tr>
        <td>1.2</td>
        <td>2.4</td>
        <td>1.19</td>
        <td>1.04</td>
        <td>0.53</td>
        <td>~467</td>
        <td>~934</td>
      </tr>
      <tr>
        <td>2.2</td>
        <td>4.4</td>
        <td>1.39</td>
        <td>1.21</td>
        <td>0.62</td>
        <td>~634</td>
        <td>~1,268</td>
      </tr>
      <tr>
        <td>3.2</td>
        <td>6.4</td>
        <td>1.50</td>
        <td>1.30</td>
        <td>0.67</td>
        <td>~740</td>
        <td>~1,480</td>
      </tr>
    </tbody>
  </table>

  <p style="margin-top: 0.75em;">
    <strong>Note:</strong> With your thrust curve dropping sharply by ~1.5 mph, increasing power mainly helps you
    accelerate and hold speed in disturbances; it doesn’t raise top speed much under this particular calibration.
    If you want, I can redo this using a geometry-based drag estimate (projected area + drag coefficients for the columns
    and bracing) instead of calibrating to the 1.5 mph point—just share the platform’s overall dimensions and how the
    columns/bracing are shaped (round/square), plus typical draft.
  </p>
</div>
```